Skip to main content

Full text of "Bulletin of the Museum of Comparative Zoology at Harvard College"

See other formats


Vfi 



;S) 



HARVARD UNIVERSITY 

Library of the 

Museum of 

Comparative Zoology 



SuL Latin OF TH 






seum o 



m 






f f;^--,-.;.;: vv 




JJ«--i^A„ 







m 



sl^f 



if«--i':i 



Volume 141 
1971 



^/^r 



HARVARD UNIVERSITY 

CAMBRIDGE, MASSACHUSETTS 02138 U.S.A. 



CONTENTS 



No. 1. West Indian Xenodontine Colubrid Snakes: Their Probable 
Origin, Phylogeny, and Zoogeography. By Vincent J. MagHo. 
December, 1970 1 

No. 2. The Milliped Family Conotylidae in North America, With a 
Description of the New Family Adritylidae ( Diplopoda : Chorde- 
umida). By William A. Shear. February, 1971 55 

No. 3. Monograph of the Cuban Genera Emoda and Ghjptemoda ( Mol- 
lusca: Ai-chaeogasti-opoda: Helicinidae). By William J. Clench 
and Morris K. Jacobson. February, 1971 99 

No. 4. The Diadematus Group of the Orb-Weaver Genus Araneus North 
of Mexico (Araneae: Araneidae). By Herbert W. Levi. February, 
1971 131 



No. 5. Evolutionary Relationships of Some South American Ground 
Tyrants. By W. John Smith and Frangois Vuilleumier. March, 
1971 : 181 



No. 6. Systematics and Natural History of the Mygalomorph Spider 
Genus Antrodioetiis and Related Genera (Araneae: Antrodiaeti- 
dae). By Frederick A. Coyle. July, 1971 269 

No. 7. A Monograph of the Genera Calidviono, Ustronia, Troschelviana, 
and Semitrochatella (Mollusca: Archaeogasti-opoda: Helicinidae). 
By William J. Clench and Morris K. Jacobson. August, 1971 403 

No. 8. The Orb Weaver Genus Neoscona in North America (Araneae: 
Araneidae). By Jonathan D. Berman and Herbert W. Levi. 
September, 1971 465 






BuLLetln OF THE 

Museum of 
Comparativ e 
oology 



-.i.,T:|-./..^i>,'ti\ -- 



West Indian Xenodontine Colubrid Snakes: 

Their Probable Origin, Phylogeny, 

and Zoogeography 



VINCENT J. MAGLIO 



HARVARD UNIVERSITY VOLUME 141, NUMBER 1 

CAMBRIDGE, MASSACHUSETTS, U.S.A. 17 DECEMBER 1970 



PUBLICATIONS ISSUED 

OR DISTRIBUTED BY THE 

MUSEUM OF COMPARATIVE ZOOLOGY 

HARVARD UNIVERSITY 

Bulletin 1863- 

Breviora 1952- 

Memoirs 1864-1938 

JoHNSONiA, Department of Mollusks, 1941- 

OccAsioNAL Papers on Mollusks, 1945- 



Other Publications. 

Bigelow, H. B., and W. C. Schroeder, 1953. Fishes of the Gulf of Maine. 
Reprint, $6.50 cloth. 

Brues, C. T., A. L. Melander, and F. M. Carpenter, 1954. Classification of 
Insects. $9.00 cloth. 

Creighton, W. S., 1950. The Ants of North America. Reprint, $10.00 cloth. 

Lyman, C. P., and A. R. Dawe (eds.), 1960. Symposimn on Natural Mam- 
malian Hibernation. $3.00 paper, $4.50 cloth. 

Peters' Check-list of Birds of the World, vols. 2-7, 9, 10, 12-15. (Price list 
on request. ) 

Turner, R. D., 1966. A Survey and Illustrated Catalogue of the Teredinidae 
( Mollusca: Bivalvia ) . $8.00 cloth. 

Whittington, H. B., and W. D. I. Rolfe (eds.), 1963. Phylogeny and Evolution 
of Crustacea. $6.75 cloth. 

Proceedings of the New England Zoological Club 1899-1948. ( Complete sets 
only. ) 

Publications of the Boston Society of Natural History. 

Publications Office 

Museum of Comparative Zoology 

Harvard University 

Cambridge, Massachusetts 02138, U. S. A. 



© The President and Fellows of Harvord College 1970. 



WEST INDIAN XENODONTINE COLUBRID SNAKES: 

THEIR PROBABLE ORIGIN, PHYLOGENY, AND ZOOGEOGRAPHY 



VINCENT J. MAGLIO 



CONTENTS 

Al)stract 1 

Introduction 1 

A^e versus Haliitus 9 

The Species Assemblages 9 

cautJierigerus species assemblage 10 

melanotus species assemblage 27 

(ludrcae species assemblage 32 

fiiiicreus species assemblage 36 

A Problematical Cenus (laltiis) 48 

Conclusions 48 

Acknowledgments 50 

Literature Cited 50 

Appendix „_ 52 



ABSTRACT 

The relationships between the thirty-three spe- 
cies of xenodontine snakes in the West Indies are 
reviewed primarily on the basis ot osteological 
and hemipenial morphology. Foiu" species assem- 
blages are recognized, distinguislied by the shape 
of the frontal and prefrontal bones and b\' the 
structure of the hemipenis. Within the caniheri- 
gcnis species assemblage three genera are recog- 
nized-A/.so/j/u's, HypsirJujnchus, and Uromacer. It 
is suggested that this group entered the West 
Indies from South or Central America, deri\ed 
from a primitive form of Alsophis. The South 
American species Alsophis chamissonis appears to 
be a relict of that primitive stock. The mainland 
genera Philoditjas and Conophis appear to be later 
specialized descendants from that same early stock. 
The three Galapagos species formerly referred to 
the genus Dromicits {hiserialis, dorsalis, and 
sk'vini), are placed in the genus AlsopJiis and 
regarded as close to the primitive mainland forms. 

The relationships of the genus laltris remain 
uncertain, but descent from West Indian Alsojiliis 
is reasonable. 

The mekinotits species assemblage has not 
progressed into the West Indies beyond the 
northern Lesser Antilles, and has almost certainly 



been derived from the mainland Leimadophis- 
Liophis-Lygophis complex. The generic name 
Dioniiciis is applied to these West Indian forms 
with the name Lcimadopliis as a junior synonym. 

Two species, andreae and parvifrons from Cuba 
and Hispaniola, respectively, share a number of 
osteological characters with Alsophis, but are like 
Dromicus externally. The hemipenis is of the 
Alsophis type and unlike that of Dromicus. Be- 
cause of their peculiar combination of characters 
these two species cannot readily be accommodated 
in any existing genus. The name AntiUopliis nov. 
gen. is proposed for them. It is suggested that 
they may be closely related to the mainland form 
Lijgophis hoursieri while the type species of that 
genus, L. lincatiis, appears to be closer to Dromi- 
cus. 

Eight species formerly placed in the genera 
Arrhytoii, Dromicus, and Darlingtonia, are con- 
sidered to form the funcrcus species assemblage. 
Except for the retention of Darlingtonia for 
hactiana, the species of this group are referred 
to the genus Arrhyton. A close relationsliip to the 
mainland genus Rhadinaca is postulated, and it is 
suggested that the two genera may have been 
derived from a connnon ancestor. The osteological 
similarities between these two groups are dis- 
cussed in terms of general seniiburrowing adapta- 
tions and are compared with other semi])urr()wiiig 
to burrowing New World colubrid snakes. It is 
concluded that these similarities represent a plnlo- 
genetic relationship rather than morpliological 
convergence. 

Four oversea colonizations from the mainland 
and numerous inter-island dispersals are recjuired 
to explain the recent West Indian fauna and its 
present distribution. 



INTRODUCTION 

Th(' \Vest Indies today contain an en- 
demic snake fauna of modest size. In the 
absence of an adequate fossil record, any 

Bull. Mus. Comp. Zool., 141(1): 1-54, December, 1970 1 



BuUctiii Miisciiiu of Conipcnalive Zoology, Vol. 141, No. 1 







Cuba 








I— 



SCALE 

100 200 
' ■ 



MILES 



'° B] 



Great Inagua 



Anguilla 

\ Antigua 
S^Cro.x\ \Guadeloupe 



Puerto 
Rico 



%°"^\ \G 

'7 jAj 




., .. , f^Dominica 
Nevisl V(. 

Montserrat/^j/ 
Marie Galante 
Martinique^'^St. 

OLucia 



Borbd^os 



Grenada, 

"^-0 Tobago 

Trinidad^ 



Fig. 1. Map of the West Indies in Mercator s projection. 



discussion of relationships within this 
group and of its history must ultimately be 
based on the inferred relationships of livinci; 
species. It is the purpose of the present 
study to examine the West Indian species 
of the subfamily Xenodontinae (sensii 
Dunn, 1928) of the family Colubridae, with 
reference to their origin, phylogeny, and 
zoogeography, so far as these can be de- 
duced from their anatomy and distribution. 
The only previous attempt to consider a 
large segment of this group was that by 
Dunn, 1932, but his work concerned only 
the Greater Antilles. Dimn relied heavily 
on the number of sensory pits present on 
each of the dorsal body scales and conse- 
quently recognized two basic generic 
groups in the West Indies; Ahophis was 
distinguished as having two pits per scale, 
and Dromicus only one pit. From these 
two groups Dunn derived all of the other 
endemic genera of the Greater Antilles. He 
also examined the dentition and hemipenis. 



concluding that these did not contradict 
his proposed relationships. However, ex- 
amination of Lesser Antillean and mainland 
species in the present stud}', as well as a 
re-evaluation of all West Indian xeno- 
dontines, does contradict these conclusions. 
Dimn weighted his characters in such a 
way that several well-defined groups of 
species remaincxl imrecognized. 

The xenodontine fauna of the \Vest 
Indies consists of the thirty-three species 
and their subspecies listed in Table 1. 
Tretanorhinus tjariahilis ssp. occurs on 
Guba, and appears to be a recent immi- 
grant from Gentral Anu^rica where se\'eral 
closely related species occur; it will not 
be considered further here. The remaining 
thirty-two species — except for ''Leima- 
doplm' mclanotus which occurs both on 
Trinidad and on th(> mainland (see Fig. 1 
for map) — are endemic to the West Indies 
and form the basis of this work. All except 
I a It lis parislii ha\'e been examined. They 



West Indian Xenodontine Colubrid Snakes 



Maglio 



Table 1 

Checklist and distribution of West Indian and some mainland^ and Galapagos^ xenodontine 
colubrid snakes. generic assignments recognized previous to this paper and recognized in the 

PRESENT STUDY ARE GIVEN FOR COMPARISON. SpECIES ARE ARRANGED IX ALPHABETICAL ORDER. 



Generic assignment 



Species 



Previous 



Present 



Distribution 



{almadensis) 

andreae andreae 

andrcac Diclopliyna 

andreae nehidatus 

andreae orientalis 

andreae ))eniu.stdae 

(an^iistilineatu-s) 

anomalus 

ater 

antillen.sis antillensis 

antillensis antiquae 

antillensis nianselli 

antillen.si.s sanetuntni 

antillensis sibonius 

( bi.seriali.s) 

callilaennis 

cantherigerus cantherigerus 

canthcrigerus adspersus 

canth erigenis h rooksi 

cantherigerus caijmamts 

cantherigerus fiiscicauda 

cantherigerus pepei 

cantherigerus ruttiji 

cantherigerus sell wartzi 

catesbyi 



( cluiniissiDiis) 

cursor 

dolichurum 

dorsulis 

dorsalis 

(dorsalis) 

exiguus cxiguus 

exiguus stahli 

exiguus subspadix 

ferox 

fremitus 

funereus 

haetiana haetiana 

liaetiana perfector 

Jul id e jidiae 

juliae co])eae 

jidiae mariae 

melanichnus 

melanotus 

ornatus 

(ixyrhynchus 

parishi 

parvifrons parvifrons 



Leimadophis 
Dromicus 



Drotiiicus 
Alsophis 
Alsopliis 
Alsophis 



Dromicus 
Dromicus 
Alsophis 



Uroniacer 



Dromicus 

Dromicus 

Arrhyton 

lultris 

Uromacer 

Dromictis 

Dromicus 



Ilypsirhynchus 
Uromacer 
Dromicus 
Darliugionia 

Dromicus 



Alsophis 
Leiniado])his 

Dromicus 
Uronuicer 
laltris 
Dromictis 



Dromicus 
Antillophis nov. gen. 



Alsophis 
Alsophis 
Alsophis 
Alsophis 



Alsophis 
Arrhyton 
Alsophis 



Uromacer 



Alsoiihis 

Dromicus 

Arrhyton 

laltris 

Uromacer 

Alsophis 

Arrhyton 



Ilyiisirhynchus 
Uromacer 
Arrhyton 
Darlingtonia 

Dromicus 



Also 1)1 1 is 
Dromicus 

Dromicus 

Uromacer 

laltris 

Antillophis nov. gen. 



Brazil 

Cuba 

Cuba 

Isle of Pines 

Cuba 

Cu])a 

Peru 

Hispaniola 

Jamaica 

Guadeloupe 

Antigua 

Montserrat 

Les Saintes 

Dominica 

Galapagos 

Jamaica 

Cuba 

Cuba 

Swan Island 

Grand Cayman 

Cayman Brae 

Cuba 

Little Cayman 

Cuba 

Hi.spaniola, Tortue 

Island, Vache Island, 

Gonave Island 

Chili, Argentina 

Martinique 

Cuba 

Hispaniola 

Gonave Island 

Galapagos 

American Virgin Islands 

Puerto Rico 

Puerto Rico 

Hispaniola 

Hispaniola, Beata Island 

Jamaica 

Hispaniola 

Hispaniola 

Dominica 

Guadeloupe 

Marie Galante 

Hispaniola 

Trinidad, Tobago, Soutli 

America 

St. Lucia 

Hispaniola 

Hispaniola 

Hispaniola 



^ Non-West Indian species are enclosed in parentheses. 



Bulletin Miisciiiti of Coinparative Zoology, Vol. 141, No. 1 



Table 1 ( Continued ) 



Generic assignment 



Species 



Previous 



Present 



Distribution 



parvifrans ciUeni 
parvifrou.s- lincolni 
IKirvifroii.s /i/^'cr 
l)(inifroiis jxiKiiii^cr 
parv if runs prat en us 
parvifron.s- rosanionde 
panifrons sii/giiis 
pdnifions tortugensis 
perfuscus 
po]\llepi.s 

])()itoriccn.sis poitoiiccnsis- 
poiioriccnsis ancgadac 
puifoiircnsis a})luinl\is 
lioitoiiccii.si.s niclmtsi 
pcrtoriccn.si-s pn/nntii.s 
purtonccnsis richardi 
portoriccnms varicgatiis 
rijger.sntdi 
nifiroitiis 

sancticntcis 
(slevini) 
(tachymenoidcs) 
tacniatiun 
variahilis 
vHtdfitni fUtatinu 
vittdttnii Idiidoi 
vudii I' lid a 

I lid a allciiiiiiis 
vudii picticep.^ 
vudii rainetji 
vudii utawanae 



Lciiiiddopliis 

Droniicu.s 

Also)}his 



Ahophis 
Alsophis 

Also])hi.s 

Droiiiicus 

Drondcus 

Arrhijton 

Trctdnorhinus 

Arrhijton 

Alsophis 



Droviicus 

Arrhijton 

Alsophis 



Alsophis 
Also I )h is 

Alsophis 

Also I )h is 

Ahophis 

Arrhyton 

(not considered) 

Arrhijton 

Alsophis 



Gonave Island 

Hispanio]a 

Hispaniola 

Hispaniola 

Hispaniola 

Vachc Island 

Hispaniola 

Tortue Island 

Barbados 

Jamaica 

Puerto Rico 

Anegadae 

Vieques 

Buck Island 

Puerto Rico 

St. Thomas 

Mona Island 

Anguilla Bank Islands 

St. Kitts, Saba, St. 

Eustatius, Nevis 

St. Croix 

Galapagos 

Peru 

Cuba 

Cuba 

Cuba 

Cu1>a 

Gri'at Baliama Bank 

Islands 

Grand Bahama Island 

Bimini Island 

Crooked Island 

Great Inagua Island 



arc divided into four species assemblages 
on the basis of skull, heniipenial, and 
external characters, and will be treated 
within thes(> groups in the following pages. 
All described subspecies except "Dioiiicus' 
amireae pen/n.sfi/ae and "D." o. meloplujrrlia 
from Cuba, "Dromicus" jiiliae copeoe from 
Guadeloupe, and "D." exi<i,uus subs})adix 
from Puerto Rico, have been examined; 
but subspecies will not be discussed further 
unless the evidence suggests a change in 
taxonomic rank. The osteology of 70 main- 



land and Galapagos species was examined 
in order to determine possible relationships 
between Antillean and mainland forms. 

Some pre\'ious generic allocations are 
here considered to be of (juestionable 
validity. I, therefore, as an initial pro- 
cedure, will disregard current generic 
assignments and use only the specific 
namc>s until probable relationships have 
been assessed and assemblages of probable 
generic \'alue can be recognized. Changes 
in nomenclature are made only where 



West Indian Xenodontine Colubrid Snakes • Maglio 5 




s soc 





Fig. 2. Skull of Alsophis conf/iengerus (MCZ 44874) showing general relationships of bones in xenodontine colubrid 
snakes. A, dorsal view; B, lateral view; C, ventral view. Left tooth-bearing elements removed. Abbreviations: bo, basioc- 
cipitcl; bs, basisphenoid; d, dentary; ec, ectopterygoid; eo, exoccipital; I, frontal; m, maxilla; n, nasal; p, parietal; pi, pala- 
tine; pm, premoxiila; po, postorbital; pr, prootic; pri, prefrontal; ps, parasphenoid; pt, pterygoid; q, quadrate; s, stapes; 
sa, surangular; sm, septomaxilla; soc, supraoccipital; st, supratemporal; tc, trabecular canal; v, vomer. Approx. X 7. 



6 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 



necessary for consistency and for the 
logical implementation of generic concepts 
as developed here. In a final section I use 
the postulated relationships to suggest a 
possible interpretation of the zoogeographic 
history of what I consider to be the valid 
groups in the West Indies. 

It is my feeling that snake genera are 
oversplit and not comparable to genera in 
other reptilian groups. This is a subjective 
judgment, however, and certainly not con- 
sistent with the philosophy and usage of 
the majority of herpetologists working on 
snakes. Since it is certainly desirable that 
taxonomic usage within the West Indies 
conform to that customary elsewhere, I 
have conserx'atively retained generic names 
(e.g., Darliniifonia) though I believe them 
to be of limited usefulness. In one case I 
have raised to generic rank a species group 
which, on current taxonomic usage, cannot 
be accommodated within any other existing 
genus. 

As Darlington (1938) and Simpson 
( 1956 ) have discussed, the islands of the 
Caribbean do not appear to haxe been 
connected with the mainland during the 
later Cenozoic and overseas migration best 
explains the available faunal evidence. 
This hypothesis seems best also to explain 
the present data. 

Characters utilized 

The choice of specific skull characters 
was made only after more than 200 skulls 
of West Indian and mainland species had 
been examined to determine which char- 
acters were least variable within a species 
and to discover which ones could therefore 
be used to infer possible relationships with 
other species. The characters so chosen 
include^ the following: 



1. The number of teeth which, in several 
of the species assemblages recognized here, 
show trends of reduction or increase from 
species to species (see Figs. 6, 20, and 29 
below and the appendix). 

2. The frontal bones (see Fig. 2 for 
labeled skull) also show considerable 
variation in shape within the West Indies, 
but prove to have distinctive proportions 
in certain assemblages, long and narrow 
on one group and nearly s(|uare in another. 
Figure 3 shows the distribution of length/ 
width ratios for the frontal bones in 29 
West Indian species. The four groups of 
species labeled A, B, C, and D represent 
the species assemblages recognized here 
on the basis of all characters combined. 
Nevertheless, even with respect to frontal 
shape alone, it is clear that assemblage C 
may be distinguished from assemblages A 
and D in having a frontal pair that is 
roughly as wide as it is long. In assem- 
blage A, most of the species have a frontal 
pair that is one and one-half to two times 
as long as wide. 

3. The shape of the prefrontal bone was 
found to be very reliable in separating 
assemblages in most cases, but (juanti- 
fication of this character was difficult. In 
Figure 4 the same 29 species as in Figure 
3 are compared with respect to the length/ 
width index for the prefrontal bone. Al- 
though assemblages B and C overlap) 
completely, both exhibit a considerably 
longer and narrower prefrontal than in 
assemblage A (Fig. 5). Although there is 
a certain degree of variation with respect 
to shape of the prefrontal and other bones, 
they remain sufficiently distinctiv^e in each 
group to be taxonomically useful. Other 
characters such as the parasphenoid width 
and skull proportions are discussed under 
each species assemblage. 



Fig. 3. Length/width indices for the frontal bone pair of 29 species of West Indian colubrid snakes. A, conther/gerus 
species assemblage; B, me/anofus species assemblage; C, lunereus species assemblage; D, andreae species assemblage. Hori- 
zontal line, observed range; vertical line, mean; open rectangle, ± one standard deviation from the mean; solid rectangle, 
95 per cent confidence interval of the mean. 



West Indian Xenodontine Colubrid Snakes • Maglio 



o 



ooo oooooooooooo 
aD0>O — cviro'tio <£>r^oo o>p — cm 



J L 



o o o o 
to ^ in to 

cvJ cvi cvj cvj cvi cj oj 

J \ I I \ \ L 




canfherigerus N=I2 



vudi! N = 6 
ater n=3 



anomalus N=2 

• melanichnus n=I 
-ig^„^ — portoricensis n=9 
-1- sancticrucis N=2 

-i- rufivenfris N=2 

1 — rijgersmai N=2 



1 

00 

I 



- antillensis N=6 
-^ ferox N = 4 

catesbyi n=7 

^i^^m^^- oxyrhynchus 

N=2 ' dor sails 

N=4 — I frenatus 



N=6 -i 



_L 



melanotus N=3 



-L perfuscus n=2 

-L- ornatus N = 2 
-L cursor N=2 
—1— yu/Zoe N = 4 
i- funereus n = 2 

. polylepis N=2 
callilaemus N=2 
— L hoetiana N=3 
— L. exiguus N=4 
vittatum N=7 
— I/, landoi N = 3 



T 
D 
1 



taeniatum n=2 



I 



andreae N=2 

parvifrons N=i3 



O 



"1 — r 



"1 — r 



T 



^ ^ 1 1 1 ^ \ \ \ \ 

OOOoOOOOOOOOOOOOOOO 

oooo— cvjro^iocot^oooo — cji*J'*;<DU) 
'———■— — -1-!_1— — evjcvicvj cjcj cjcj 



8 



BuUctin Museum of Comparative Zoolo'^ij, Vol. 141, No. 1 



o 



o 

00 



o 
o 



o 



o 



o 



o 

00 



o 
q 

cvj 

I 



o 

CVJ 
cvj 



O 
^. 

CVJ 

I 



O 

cvi 

I 



o 

00 
CVJ 



o 
o 

ro 



O 

CVJ 

to 



o 

to 



o 

ro' 




cantherigerus N=I2 
vudii N=6 
I ater n=3 
. onomalus N-Z 

• melanichnus N=l 
portoricensis N=7 
sancticrucis N=2 
-J- rufivenfrls N=2 
n'jgersmai N = 2 

antillensis N = 6 



/erox N=3 



cafesbyi N = 7 



I 



00 



1 



oxyrhynchus N = 5 

-L dorsal is N = 2 

-L frenatus N=4 



melanotus N=3 



-J — perfuscus N = 2 



-I — ornafus N=2 
I cursor N = 2 



J — ;'u//oe N-4 



± 



funereus N=2 



— ^ — polylepis N = 2 
callilaemus N = 2 



haefiana n=| 



± 



ex/guus N = 4 



V landoi N = 3 



P — vitfatum 
N = 7 



. taeniatum N=2 



T 














1 




3 




D 






nnruifmnt: M-l'X 






1 1 1 




1 












1 

O 
to 


1 
o 

00 


1 

o 
o 

• 


1 

o 

CVJ 


1 

o 

• 


1 
o 


1.80- 
2.00- 


T - 

o 

CVJ 
cvj 


2.40- 
2.60- 

2.80- 

3.00- 
3.20- 


— 1 — 
o 

lO 


I - 
o 

to 



Fig. 4. Length/width indices for the prefrontal bone of 29 species of West Indian colubrid snakes. Symbols as in Fig. 3. 



West Indian Xenodontine Colubrid Snakes 



Maglio 



The discussions of hemipenial morphol- 
og>^ are based on dissections in situ of the 
une verted organ. Terminology is after 
Dowhng and Savage (1960). 

It must be emphasized that in this study 
most measured parameters show various 
degrees of overlap in range for various 
species. This does not, however, lessen 
their value in attempting to recognize 
phylogenetic relationships through an 
analysis of moiphological similarities. If 
an insular series of closely related species 
has been successively derived by island- 
hopping in a linear sequence, we might 
expect any two adjacent forms to show a 
greater similarity, barring extreme special- 
ization, than the two geographically 
terminal species of the series. 

The characters used here are those that 
combine relatively little intraspecific vari- 
ability and enough variation between spe- 
cies to be useful in the study of intrageneric 
relationships. Certain characters such as an 
apical awn on the hemipenis, the shape of 
the prefrontal bone, or certain skull pro- 
portions appear to be constant within 
groups of species that show a close relation- 
ship in totality of characters combined, and 
are therefore considered to be of maximum 
value in indicating true affinities. Others, 
such as the number of pits on each dorsal 
body scale, have sometimes proved incon- 
sistent \\'ith the majority of other traits 
and have therefore been considered un- 
reliable at the le\'el of major groups. ^ 

AGE VERSUS HABITUS 

When attempting to establish relation- 
ships between extant forms based entirely 
on morphological considerations, it must 
be reasonably certain that differences are 
not the result of allometric phenomena. 
Likewise, it must be reasonably clear that 
apparent similarities are not the result of 
habitus rather than a close phylogenetic 
relationship. 

Juvenile specimens, when available for 
comparison with the corresponding adult 



forms, show a number of consistent differ- 
ences in the structure of the skull which 
appear to be related solely to size. The 
major differences may be summarized as 
follows : 

Juvenile skull 

Skull relatively wide 

Quadrate thin and triangular 

Supratemporal relatively short 

Crests low and rounded 

Postorbital small, non-projecting 

Orbital foramen very large 

Pterygoids short, not projecting beyond 

foramen magnum 
Maxilla relatively lightly built 
Bones of brain case thin 

Adult skull 

Skull relatively narrow 

Quadrate with rodlike shaft 

Supratemporal relati\'ely long 

Crests high and sharp 

Postorbital large, projecting 

Orbital foramen small 

Pterygoids long, projecting far beyond 

foramen magnum 
Maxilla relatively massive 
Bones of brain case thick 

It is clear that these characters should 
not ordinarily be given high taxonomic 
weight unless the comparison is between 
two species of comparable adult size. 

Similarities resulting from habitus adap- 
tations are more difficult to establish be- 
cause the mode of life of these species is 
so poorly understood, and also because the 
adaptive significance of certain characters, 
such as scale pits, is at present unknown. 
Character convergence resulting from 
habitus similarity can, however, be inferred 
if a large number of characters are studied 
together. This point will be discussed in 
greater detail under the funereus species 
assemblage below. 

THE SPECIES ASSEMBLAGES 

Four species assemblages have been dis- 
tinguished among the 32 species of West 
Indian xenodontine snakes here under 
study. These are characterized on the 
basis of a number of traits as follows: 



10 



Build ill Museum of Cotu para five Zoology, Vol. 141, No. 1 



Prefrontal 


Frontals 


Hcmipenis 


Size 


long anteroposteriorly 


long and narrow 


no disk 


large 


short anteroposteriorly 


short and narrow 


apical disk 


medium 


short anteroposteriorly 


long and narrow 


no disk 


medium 


short anteroposteriorly 


square 


no disk 


small 



cantherigerus 

assem])lage 
inelanotus 

assemblage 
undrcae 

assemblage 
funereus 

assemblage 



CANTHERIGERUS SPECIES ASSEMBLAGE 

Included West Indian species: anomalus 
Peters, Hispaniola; antillensis ( = leu- 
comdasY Schlegel, Antigua, Montserrat, 
Guadeloupe, Les Saintes, Dominica; ater 
GossE, Jamaica; cantherigerus- Bibhon, 
Cuba, Isle of Pines, Swan Island, Grand 
Cayman, Little Cayman, Cayman Brae; 
cateshiji Schlegel, Hispaniola, Tortue Is- 
land, Vache Island, Gonave; dor.salis Dunn, 
Gonave; ferox Gunther, Hispaniola; fren- 
atus GiJNTHER, Hispaniola; mehnichnus 
Cope, Hispaniola; oxyrliynchus Dumeril 
AND BiBRON, Hispaniola; portoricensis Rein- 

HARDT AND LuTKEN, PucrtO RicO, Moua 

Island, Virgin Islands; rijgersmm Cope, 
Anguilla Bank Islands; rufiventris Dumeril 
AND BiBRON, Saba, St. Eustatius, St. Kitts, 
Nevis; .sancticrucis Cope, St. Croix; viidii 
Cope, Great Bahama Bank Islands, Great 
Inagua. 

Osteology. The present group of species 
may be distinguished from other West 
Indian xenodontincs by a number of skull 
features, the most characteristic of which 
is the shape of the prefrontal bone (see 
Fig. 5). Here this element is wide antero- 
posteriorly with a broad and strongly con- 
vex anterior edge. The lower margin of 
this anterior edge cun^es medially aboxe 
the lacrimal foramen, so that the latter 
opens anteroventrally. 



^As discussed by Schwartz (1966: 178), 
Brongersnia's (1937) analysis of Schlegel's co- 
types and his choice of the Guadaloupe-like speci- 
mens as the lectotype of Ahophis antillensis have 
reduced the name leucomclas to the junior 
synonymy of antillensis. 

- Senior synonym of angulifer; see Smith and 
Grant, 1958. 



Within the assemblage, the species 
cantherigerus (Cuba) has the lowest num- 
ber of teeth, with an average dental 
formula of about 12 -t- 2 maxillary, 10 
palatine, 26 pterygoid, and 19 dentary teeth 
(see Fig. 6 and the Appendix for vari- 
ation). The skull (Fig. 7) is long and the 
cranium is moderately deep dorsoventrally. 
The frontals are widest anteriorly where 
they make contact with the prefrontals and 
are strongly emarginated above the orbits. 
A short, stout postorbital bone is articulated 
in a deep notch on the parietal bone in 
such a way that a prominent flange or 
lateral extension of the parietal intervenes 
between the postorbital and the frontal 
bones (see Fig. 2). A weak, but clearly 
visible groove marks the dorsal midsagittal 
line of the parietal bone. The parasphe- 
noid, forming the midventral surface of 
the skull, is narrow and has a deep trabec- 
ular canal on each lateral surface. This 
groove extends from the orbital foramen 
to the nasal capsule. A dorsal extension 
of the parasphenoid bone above the 
trabecular canals separatees the two orbits, 
forming a thin intcrorbital partition.' The 
supratemporal is strong and curved, and 
extends some distance beyond the occiput. 
The cjuadrate is long and straight. 

The species vudii on the islands of the 
Great Bahama Bank does not significantly 
differ ostcologically from cantherigerus. 
The dental formula is about 12 + 2 maxil- 
lary, 10 palatine, 24 pterygoid, and 21 
dentary teeth for vudii vudii and is roughly 
comparable to that of cantherigerus. A 
peculiar feature of vudii is the melanic 



^ Equals frontal crests of Underwood, 1967. 



West Indian Xenodontine Colubrid Snakes • Maglio 11 









Fig. 5. Comparison of the three prefrontal bone types 
found in the four species assemblages of West Indian 
colubrid snakes as discussed in the text. A, conther/gerus 
assemblage type [Ahophis confher/gerus, MCZ 11200); B, 
me/onofus and andreae assemblage type (Antillophis parvi- 
frons, nov. gen., MCZ 77227); C, funereus assemblage type 
{Arrhyton polylepls, MCZ 81020). For each: left, lateral 
view; right, anterior view. Not to scale. Approx. X 10. 



tissue lining the cranial cavity of every 
specimen examined. This tissue imparts 
a bluish gray color to the skull. A similar 
condition occurs occasionally in cantheri- 
genis, and also in cateslnji, dorsalis, 
frenatus, and oxijrhijnchus where it is the 
usual condition. 

Although only two specimens including 



the type were available for study, vudii 
utowanoe from Great Inagua differs from 
all the subspecies of vudii on the Great 
Bahama Bank in several characters. The 
nasal bone is distinctive in shape, but this 
character appears to be somewhat more 
variable than most skull characters and is 
therefore of uncertain significance. In its 
dentition, however, vudii utowanae is quite 
distinct from the other subspecies. The 
dental formula is about 15 + 2 maxillary, 
13 palatine, 30 pterygoid, and 23 dentaiy 
teeth, and is greater for every tooth-bearing 
element. When additional specimens of 
utowanae become available, a greater de- 
gree of overlap with the Bahama Bank 
forms may become evident. However, 
utowanae still will largely lie outside the 
neatly clustered range for the other sub- 
species of vudii. As in vudii vudii, the skull 
of utouanae appears bluish gray due to the 
melanic tissue lining the cranial cavity. 

On Jamaica the species ater has a dental 
formula higher than that of eantJierigerus. 
The skull is generally flatter (Fig. 8) and, 
as a result of this flattening, the nasal bones 
are closer to the septomaxilla, and the 
frontal bones touch the trabecular canals 
ventrally so that the interorbital partition is 
very small, consisting only of that portion 
of the parasphenoid bearing the trabecular 
canals. The frontal bones are relatively 
shorter and wider (Fig. 9) in contrast to 
the long, narrow frontals of the Cuban 
species. In all other skull characters the 
two forms are very similar. The septo- 
maxillae are expanded anteriorly and widen 
immediatelv behind the premaxilla (Fig. 
lOB). 

The name capistrata, introduced by 
Gosse (1851: 373) for a patterned form 
from Jamaica, was synonymized with ater 
by Boulenger (1894) without comment. 
Two specimens of this form from the 
British Museum were made available to 
me for comparative puiposes. Both are 
smaller than typical ater and differ from 
it in a number of osteological characters 
which in other species are related to 



12 Bulletin Miisciiiii of Coiujniiafivc Zoolofnj, Vol. 141, No. 1 



MAXILLARY TEETH 
— cvjrO'j-mtflr-oooo— c^ff' 

— — — fsj OJtMCVJ 



1 I I I 



I I I \ I L 



canfherigerus N = I2 



— — vudii N = 4 

• vudii utowanae N=2 

• ater N = 4 
• anomalus N-2 

• melanichnus N-l 
^—^—. portoricensis N=IO 

^— sancticrucis N=2 

— rufiventris N=2 

— rijgersmai N-2 

antillensis N=6 

— ferox N=3 



catesbyi N = 6 

oxyrhynchus N-7 

— dorsalis N=2 
frenatus N = 4 

1 — I — \ — rT~i — r~\ — n i i r~ 

— cjro^intON-oooO — cMto 

CVJOJOJM 



PTERYGOID TEETH 

f^ooo^O — cjro^iou)h-00CT)O — cvj 
cJcJcvjc\Jc\JcMc\Jc\Jc\JOJroroio 

I I I I I I \ I I I I I I I I I 



vudii 



canfherigerus 



vudii utowanae 



— ater 



• anomalus 



• melanichnus 
portoricensis 

sancticrucis 



rufiventris 

rijgersmai 

antillensis 



ferox 



-— catesbyi 
oxyrhynchus 



• dorsalis 
— frenatus 



I I I I I I I I I I I I 
h-oo<T>0— cvjrO'^-ioUJt^ajaiO— cj 
cvJOJCvjoJCvJCNJOJoJcocvjfOtOro 



TT 



PALATINE TEETH 



N CO CT) 



O — cviro^ioUJh- 

_l I \ I I I I I 

canfherigerus 



— vudii 

— vudii utowanae 
ater 

— anomalus 

• melanichnus 

portoricensis 

— sancticrucis 
rufiventris 



rijgersmai 

antillensis 

• ferox 

— catesbyi 

oxyrhynchus 

. dorsalis 
. frenatus 

1 — I — rn — rn — i i i i — i — 
r~cocn2z^[2i:2^I^ 



DENTARY TEETH 

cDf^ooo^O— cvJro^incor-ooo^O — ojro^io 
cvjc\JCVJcJojcocvJCMoJcvjror<^rOrOrotO 

I I I \ \ l_J I 1 I I I 1_J \ \ \ \ 1 L_ 



canfherigerus 
— vudii 

— vudii utowanae 

ater 



— anomalus 



melanichnus 



portorice nsis 



— sancticrucis 
rufiventris 



— rijgersmai 
antillensis 



— ferox 



catesbyi 
oxyrhynchus 



•^— dorsalis 

• frenatus 

~\ I I I I \ — \ — \ — \ — \ — \ — \ — \ — \ — \ — \ — \ — I — I — r 
lOf^oooiO — cviro'^u^tDf^coo o — cjrO'j-io 
oJOJcvJCMCJOJOJcdOJCO rorOfOtOfOro 



Fig. 6. Observed ranges of variation in numbers of teeth on each tooth-bearing element for the 15 West Indian species 
of the contherigerus species assemblage. 



ontogenetic changes. These inehide rela- capistroto represents a juvenile stage ofj 

tively narrower frontals, a broad rounded ater, and we may follow Boulenger in 

cranium, low crests and ridges, and thin rc\garding the two as synonymous, 

cranial bones. It thus seems likely that On Hispaniola there are eight species 



West Indian Xenodontine Colubrid Snakes • Maglio 13 






5 mm 



Fig. 7. Skull structure of Alsophis cantherigerus, MCZ 56429. A, dorsal view; B, lateral view; C, ventral view. 



that may be placed in the present species 28 pteiygoid, and 24 dentary teeth. The 

assemblage. In the rare form melanichmis\ posterior processes of the vomer are later- 

the numbers of teeth are higher than in ally expanded into flat plates, oval in shape 

either cantherigerus or ater- the dental when viewed from below. This character 

formula is 18 + 2 maxillary, 16 palatine, is not seen in any other West Indian 



14 Bulletin Museum of Conipamtive Zoology, Vol. 141, No. 1 




oooooooooooo 
r^oooo — coroT^intDN-QO 



J III' I I I I \ L^ 

cantherigerus N = I2 
vudii N = 6 

^ ater N = 3 

— I — anomalus N = 2 
. melanichnus N= I 

-i^^^a^ — porforicensis N=7 

— ' — sancticrucis N = 2 
— I — rufiventris N=2 
— I— rijgersmoi N=2 



-• — fez-ox N = 3 
cafesbyi N = 7 
oxyrhynchus N=6 
-L- dorsalis N = 2 
J frenatus N = 4 



anfillensis 
N=6 




1 
o 


■f- 
O 


■~r- 
o 


1 
o 


■■I — 
o 


1 
o 


1 

o 


1 
O 


1 

o 


1 

o 


1 
o 


1 
o 


N- 


CO 


en 


o 




<M 


ro 


'J- 


ID 


to 


r^ 


CD 



Fig. 8. Frontal bone width/preorbital skull depth indices for the 15 West Indian species of the conther/gerus species 

assemblage. Horizontal line, observed range; vertical line, mean; open rectangle, it one standard deviation from the 
mean; solid rectangle, 95 per cent confidence interval of the mean. 

xenodontine. In comparison to canthcri- (Fig. 11). The skull has numerous high 

gerus, the postorbital bone is wider in crests and ridges for muscle attachment, 

proportion to its length, and the frontal is but these crests appear to develop with 

relatively slightly shorter. The supra- positive allometry in most large speci- 

temporal is short and stout. mens of xenodontines and are almost 

Another Hispaniolan species, anomalus, certainly related to the great size of this 

is closer in many ways to cantherigerus species. In its dentition, anomalus shows 

(Cuba) than it is to melanichnus, espe- only slight modifications from the con- 

cially in the structure of the vomer, post- dition found in cantherigerus (Cuba), 

orbital, and supratcmporal. The skull is and has a formula of about 12 +2 maxil- 

proportionately slightly wider and more lar\% 8 palatine, 20 pterygoid, and 19 

dorsoventrally depressed, anterior to the dentary teeth. 

orbit, than in the Cuban form. The pre- Also found on Hispaniola is the well- 
maxilla is a solid, heavy structure, semi- defined species ferox. (I include speci- 
circular in ventral view, and quite unlike mens from the southwest peninsula of 
that of other members of the assemblage. Haiti which consistently lack a loreal 
Ventrally, the parasphenoid is wider be- scale.) This form is remarkably like ater 
neath the orbits than in cantherigerus and (Jamaica) in dorsoventral flattening of 
forms only a short interorbital partition the skull, in the short, wide frontal, and 



West Indian Xenodontine Colubrid Snakes • Maglio 15 



o 
o 



oooo OOOOOOo ooooo 
— cMrO'^ ir)CO>-oocDO — c\Jro^tn<X) 

~~~— — — — — — CMCVJCVICVJcMCJCVJ 

-J ^ 1 1 1 \ \ 1 \ I I 




cantherigerus N = 12 



vudii N=6 
afer N=3 



anomalus N = 2 

• melanichnus N = l 

r 



portoricensis N = 9 
-J — sancficrucis N=2 

— 1— rufiventris M-Z 

1 rijgersmoi N=2 




antillensis N=6 
ferox N=3 



catesbyi N=7 



a^— oxyrhynchus N=6 

-■ dorsal is N = 2 

frenatus N=4 



o o 
o - 



o 

CVJ 



o 

rO 



o 



o 

in 



O 



o 



o 

00 



o 



o 
o 



— CM 



o o 

— C\J 



CJ 



CJ 



o 
to 

CM 



o 

CM 



O 

in 

CM 



O 
CM 



Fig. 9. Length /width indices for the frontal bone pair of the 15 West Indian species of the can/her/gerus species as- 
semblage. Symbols as in Fig. 8. 



in the veiy small parasphenoid inter- 
orbital partition. The septomaxilla is even 
more expanded than in the Jamaica spe- 
cies (Fig. IOC), and the nasal area is 
depressed dorsoventrally as in that species 
so that the nasal bones lie close to the 
septomaxilla. In all of these characters 
this species is suggestive of catesbyi (dis- 
cussed below). In its dental formula ferox 
sho^^s a reduction in the number of teeth, 
as compared with atcr. The teeth of ferox 
are larger than in oter, but this character 
is somewhat variable in specimens of 
equal size. In all other characters the 
similarity between ferox and ater is strik- 
ing. The only notable osteological differ- 
ence is in the shape of the nasal (Fig. 12), 
which in ferox is wide anteriorly and 
tapers off behind. 



The four remaining Hispaniolan species 
are clearly united into a single sub- 
assemblage on the basis of external 
moiphology (see below). Within the sub- 
assemblage, catesbyi is the least special- 
ized and is very similar to ferox in skull 
structure; few cranial characters can 
adequately separate the two species. The 
major difference is in the higher dental 
formula in catesbyi. The teeth are en- 
larged as compared with ater, but not as 
much as in ferox, except for the palatine 
and pterygoid teeth which are as large 
as in that species. The most notable 
similarities between ferox and catesbyi are 
in the anteriorly expanded septomaxilla 
(Fig. 10), and in the dorsoventrally de- 
pressed preorbital portion of the skull. The 
nasal of catesbyi, although distinct in 



16 Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 







Fig. 10. Left septomaxilla of four species of tfie can- 
therigerus species assemblage. Dorsal view. A, Ahophis 
con/herigerus, MCZ 8611; B, A. ater, MCZ 6005; C, Hyp- 
sirhynchus ferox, MCZ 64785; D, Uromacer catesbyi, MCZ 
3605. Not to scale. Approx. X 5. 



shape, is not very different from that of 
ferox. The more highly speciahzed species 
of this subassemblage, oxyrhijncluis, dor- 
salis, and frenatus, have greatly elongated 
nasals, vomers, and septomaxillae. The 
dentary, maxilla, and palatine are also 
elongated, thus extending the dental row 
forward. The species oxijrhynchus and 
(lorsalis are the most highly specialized 
forms \\ith an exaggeration of all these 
characters. 

On Puerto Rico and the Virgin Islands 
the species portoricensis occurs. In the 



structure of its nasal, premaxilla, supra- 
temporal, and (juadrate (Fig. 13), it is 
more like canthevigevus (Cuba) and anO' 
mains (Ilispaniola) than like melanichmis 
(Hispaniola). The interorbital partition is 
small and the frontal is shorter antero- 
posteriorly than in the Cuban species. 
Ventrally the parasphenoid is wider than 
in either cantherigenis or mehnichnus. The 
dental formula is higher than in cantheri- 
genis, but there are fewer maxillary and 
palatine teeth than in mehnichnus. The 
subspecies portoricensis nicholsi and p. 
anegadae, both from the \'irgin Islands, are 
similar to each other in having a longer 
and narrower frontal bone than other sub- 
species of portoricensis. In p. anegadae the 
parasphenoid is very narrow, but in p. 
nicholsi this bone is as wide as in other 
subspecies of portoricensis. 

On the northern islands of the Lesser 
Antilles there are four species that are 
osteologically very close to portoricensis. 
Most similar is sancticrucis from St. Croix 
which is osteologically inseparable from 
that species. On Saba, St. Kitts, St. Eusta- 
tius, and Nevis, the species riifiventris is 
also very close to the Puerto Rican species 
in most osteological characters. The form 
on St. Barthelemy and Anguilla, rijgersnjai, 
is another member of what may be called 
the portoricensis species group. In most 
characters of the skull and in the dentition, 
this form is like the Puerto Rican species. 
The nasal is like that of riifiventris, Ixit the 
skull is narrower throughout, and the pre- 
maxilla reduced in size. The frontal is 
proportionately narrower than in other 
members of this subgroup. One other 
member of the group is antillensis ( = leii- 
comehis) which in most characters of skull 
moiphology is like riifiventris (St. Kitts, 
etc.). Unlike rijgersmai the frontal is broad 
and the premaxilla is unreduced. The 
dentition in all of these species of the 
portoricensis species group is similar when 
the degree of \'ariation in each is taken into 
account (see Fig. 6). 

The species of the portoricensis species 



West Indian Xenodontine Colubrid Snakes • Maglio 17 



oooooooooooooo 

CDOCVI^tDOOOCVJ^J-vOOOOCVI^ 



cvJcvJcMcvJcvjrorOro 

=i= L 1 I \ I I ■ 



.cantherigerus N=ll 



vudii N=5 



afer N=3 
anomalus N = 3 



• melanichnus N = 

^^m^mimimm^ portoricensis N = 7 

-< sancticrucis N=2 

—I— rufivenfris N = 3 

— I — rijgersmai N=2 
anti lien sis N = 6 
— ^erox N = 3 



catesbyi N = 7 



Hii^^^iaA- oxyrhynchus N = 6 
dorsal is N = 2 
' frenafus N = 4 



~T 1 ^ 1 1 1 1 1 1 1 1 1 \ r 

oooooooooooooo 

OOOCJ'tCOOOOcvJ^UJOOOCVJ^ 

— — — — — cvJcvjcJcvJCJrOfOfO 



Fig. 11. Interorbital height/ventral width indices for the parasphenoid bone of the 15 West Indian species of the con- 
fherigerus species assemblage. Symbols as in Fig. 8. 

group show a far closer relationship among other in most external characters. Several 

themselves than they do to any other spe- forms such as oter (Jamaica) and the ferox 

cies in the cantherigerus assemblage, al- populations from the southwest peninsula 

though they are clearly part of that of Haiti (for which the name scaloris Cope 

assemblage. The rather minor morpho- is available) have lost the loreal scale, but 

logical distinctions which may be used to this has occurred repeatedly in many 

distinguish these species are shown dia- groups and cannot be considered more than 

grammatically in Figure 14 in relation to a specific or subspecific difference. All the 

their geographic distribution. These five members of this group are similar in size, 

well-defined taxa may represent no more most falling between SOO and 1300 mm in 

than geographic races within a moipho- total body length. There are al\\'ays eight 

logically variable species, but until addi- supralabials, usually with the third, fourth, 

tional data are available it is best to retain and fifth entering the orbit. The number 

these forms as distinct species. of dorsal scale rows ranges from 17 to 23. 

Extcrnnl morplwJogy. Except for the Ventral scale counts range approximately 

specialized semiarboreal forms on Hispan- between 160 and 230, and caudal counts 

iola, the members of the cantherigerus roughly between 100 and 150. The anal 

species assemblage are very similar to each plate is usually divided, but may be 



18 Bulletin Musciini of Comparative Zoology, Vol. 141, No. 1 



taeniatum 
MCZ 19844 



funereus 
MCZ 44901 



calliiaemus 
MCZ 69080 



haetiana 
MCZ 65105 



exiguus 
MCZ 37356 







chamissonis 
MCZ 6510 



cantherigerus 
MCZ 44878 



melanichnus 
MCZ 7836 



portoricensis 
MCZ 46503 



ruflventris 

MCZ 6130 








ferox 
MCZ 37665 



cafesbyi 
MCZ 13676 



oxyrhynchus 
MCZ 13768 



anomalus 
MCZ 12644 



I. dor sails 
MCZ 25561 








melanotus 
MCZ 49024 



cursor 
MCZ 6011 



juliae 
MCZ 6138 



parvifrons 
MCZ 77228 



andreae 
MCZ III57 








Fig. 12. Shape of the nasal bones for representative species of the four species assemblages of West Indian colubrid 
snakes as discussed in the text. Not to scale. Approx. X 7. 



West Indian Xenodontine Colubrid Snakes • Maglio 19 






5mm 



I — I — ' — ' — I — t 



Fig. 13. Skull structure of Ahopbis portoricensis, MCZ 46503. A, dorsal view; B, lateral view; C, ventral view. 



single in some specimens of sancticrucis 
(Schwartz, 1966). The Hispaniolan species 
ferox has a single apical pit on the dorsal 
body scales, whereas cateshyi, oxyrlu/nchus, 
dorsalis, and frenatus have none. The re- 
maining species have two pits. 

The species ferox differs externally from 
other members of the assemblage. The 
snout is elongated, the rostral scale forms 



an acute angle with the top of the head, 
and the eye is large and bulging. The 
pupil shape in ferox is usually oval, but 
out of 28 preserved specimens from the 
population, 26 had round or 
rounded pupils, and only one 
had a distinctly oval pupil in 



southwest 
irregularly 
individual 
both eyes. 
The 



arboreal species of Hispaniola, 



20 



Bulletin MusciDu uf Cumpaiativc Zoology, Vol. 141, No. 1 



ST. THOMAS 
portoricensis 




SABA ETC 

rufiventris 

frontal broad 
premaxilla unreduced 
nasal modified 



ST. CROIX 
sancticrucis 

frontal broad 
premaxillcry unreduced 



ANGUILLA 

rijgersmoi 

frontal narrow 
premaxilla reduced 
nasal modified 



GUADELOUPE ETC. 

antillensis 
frontal broad 
premaxilla unreduced 
nasal modified 



Fig. 14. Geographic distribution of several morphological characters within the porfor/censis subgroup of the contfierig- 
erus species assemblage. Circle represents the supposed center of dispersion of this group. 



cateslnji, oxyrJiyncJtus, dorsaJis, and fre- 
natiis, are clearly united in a well-defined 
subgroup. All, except cates])yi, are long, 
slender, and clearly adapted for tree living. 
A recent study by Horn ( 1969 ) demon- 
strates the specific synonymy of "wetmorei" 
with frenatus and of ^'scondax" with 
cateshyi. 

The subspecies vudii titowonae as de- 
scribed by Barbour and Shreve (193S) has 
a higher ventral and subcaudal scale count 
than other races of vudii. 

Ileinipenis. The structure of the hemi- 
p)enis is remarkably uniform within the 
C(intherig,erus assemblage. The sulcus sper- 
maticus is divided near the base of the 
organ and each branch extends to the tip 
of one lobe of the deeply bifurcated apex 
(Fig. 15). Several rows of longitudinally 
arranged stout spines are present along the 
middle one half of its length; these grade 
into numerous smaller spines basally. The 



base may also hv nude or ha\'e long plicae. 
The sulcus is bordered by a fringe of folded 
tissue bearing modest sized, closely spaced 
spines that grade into smaller ones distally. 
On the apex of each lobe, a reticulate net- 
work of tissue surrounds the sulcus, with 
moderate to long filiform papillae (papil- 
late calyces). These calyces usually grade 
into the surrounding tissue proximally, but 
may form a more or less well-defined 
capitate structure as in protoricensis, 
rij'^crsmai, and antillensis. The hemipenis 
in ferox does not differ from that of other 
species in this group except in having 
longer apical papillae. In cateslyyi and the 
other arboreal snakes of Hispaniola the 
hemipenis is proportionately shorter than, 
but basically similar to, that of cantheri- 
iS.erus. It is more heavily spinose, and the 
apical papillae on the sulcate surface ex- 
tend further basally to the point of division 
of the sulcus. 



West Indian Xenodontine Colubrid Snakes • MagJio 21 





Fig. 15. Hemlpenis morphology In the cantherigerus species assemblage; semidiagrammotic. Organ is 
dissected in situ. A, Alsophis ater, MCZ 6005; 6, Alsophis portoricensis, MCZ 58804. Approx. X 5- 



uneverted and 



Mainland relationships and origin of the 
group. Examination of South and Central 
American genera has revealed several that 
are morphologically very close to the 
cantherigerus species assemblage. Of the 
four mainland species customarily referred 
to the genus ^'Dromicus," three are clearly 
related to the present group; these are 
chamissonis, taclupnenoides, and angustili- 
neatus. The fourth species, '^Dromicus" 
amazonicus, is allied to the melanotus spe- 
cies group to be discussed below. 

The species taclupnenoides (Peru) and 
angustiUneatus (Peru) differ from chamis- 



sonis (Chile and Argentina) only in several 
minor characters, and what is said below 
about the latter will apply also to these 
two species. The major distinction between 
elunnissonis and cantherigerus (Cuba) is 
the lower number of teeth in the former 
( see appendix ) ; in all other skull characters 
the two are extremely close. Externally 
there are no differences which would argue 
against a close relationship, although the 
mainland species have only one pit on each 
dorsal body scale. In the structure of the 
hemipenis the sulcus spermaticus is less 
deeply divided in chamissonis, but the 



22 Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 





Fig. 16. Premaxilla in a typical West Indian species of 
the confher/gerus species assemblage compared with that 
bone characteristic of the Galapagos species of this group. 
A, Alsophii conther/gerus, MCZ 8611; B, Akophis slevini, 
MCZ 28470. Ventral view. Approx. X 12. 



arrangement of spines and the nature of 
the apical differentiation is similar to that 
of the Cuban form. As a whole, chamis- 
sonis must be considered as a mainland 
representative of the canthcrigerus assem- 
blage. 

Two other mainland genera showing a 
close relationship to the contlierigerus 
group are Fhilodnjas from South America 
and Conophis from Central America. Osteo- 
logically these genera are very close to 
chamissonis, except that both have well- 
developed groo\'es on the posterior maxil- 
lary teeth. Externally Fhilodnjas may have 
one or two apical pits per dorsal body scale, 
and Conophis has none, but in size, scale 
pattern, and scale count they do not differ 
significantly from chamissonis. The struc- 
ture of the hemipenis in both genera is of 
the cantherigerus type, with a deeply 
forked sulcus, lateral spines, and an apical 
ornament of papillate calyces arranged as 
in that group. 

The present distribution of chamissonis, 



tacliymenoides, and angustilineatus makes 
it difficult to derive any part of the 
West Indian fauna from them unless we 
postulate a former widespread distribution 
for an ancestral group. Fhilodnjas and 
Conophis now occupy the geographic areas 
which the ancestral genus must have in- 
habited if the West Indian representatives 
of this group were derived by waif dis- 
persal from the mainland. However, the 
presence of rear fangs in these two genera 
would appear to preclude them from any 
direct ancestry to the nonfanged Antillean 
group. The morphological relationships be- 
tween Fhilodnjas, Conophis, and chamis- 
sonis, etc., as well as their present 
distributions, suggest a widespread ances- 
tral group common to all three. The species 
chamissonis, tacliymenoides, and angiistili- 
7ieatus may represent relatively undifferen- 
tiated relicts of that ancestral group, and 
Fhilodnjas and Conophis specialized rear- 
fanged descendants. The development of 
rear fangs has occurred repeatedly in a 
number of unrelated groups and is no bar 
to the relationship here suggested. If such 
a specialization was of selective advantage 
over the nonfanged condition, we might 
expect these forms to haxe displaced the 
ancestral type from much of its former 
range. 

Taxonomy. The cantherigenis species 
assemblage as defined above on the basis 
of osteological and hemipcMiial characters 
may be expanded to include chamissonis, 
taclnjjnenoides, and angustilineatus from 
South America. The three species from the 
Calapagos referred to the genus Dromi- 
ciis" must also be placed here. These three 
species-biserialis, dorsalis, and slevini-are 
very much like the present assemblage in 
external characters, osteology, and hemi- 
penial moiphology. However, they are 
clearly closer to each other, and represent 
products of speciation on the Galapagos. 
The dental formula is similar in all three 
forms and is as low as in their mainland 
relatives (appendix). A minor but distinc- 
ti\'e character which sets these forms apart 



West Indian Xenodontine Colubrid Snakes • Maglio 23 



CENTRAL AMERICA 

Conophis 
no scale pits 
rear fangs 



GALAPAGOS 

Also phis 
2 scale pits"^ 



HISPANIOLA 

Hypsirhynchus Uromacer 

I scale pit no scale pits 

elliptical pupil arboreal specializations 




NDIES 



Alsophis 
^2 scale pits 



HISPANIOLA 

laltris 

2 scale pits 
rear fangs 

Fig. 17. Geographic distribution of several morphological characters in the genus Ahophis and in two related mainland 
genera. The circle represents the supposed center of dispersion for this group. The I symbol indicates that the ancestral 
form of Alsophis is extinct on the mainland. 



from other members of the cantherigerus 
group is the shape of the premaxilla (Fig. 
16); here this element is depressed an- 
teriorly in the midline so that the lateral 
processes appear to eurve forward and then 
back in a winglike fashion. 

Within the West Indies as discussed 
above, utoiranae from Great Inagua is 
somewhat distinct from other subspecies of 
vudii. At the time of writing only two 
specimens including the type were avail- 
able for study. Although the differences 
of these two specimens from vtidii viidii 
appear to be more than subspecific, it 
seems advisable to retain iitowanae as a 
subspecies of vitdii until additional speci- 
mens are available and the range of vari- 
ation better known. 

The four arboreal species on Ilispaniola 
~cates]}yi, oxyrhijnchus, dorscdis, and fre- 
natus-aie uncjuestionably closely related 
and united morphologically; they form the 
well-defined genus Uromacer. The Hispan- 



iolaii species ferox is osteologically undif- 
ferentiated from other members of the 
group except for its enlarged teeth, and in 
many ways it is intermediate between ater 
(Jamaica) and Uromacer cateslyyi. Mertens 
(1939) favored retention of ferox in a 
distinct genus (Hypsirliyncliiis) because of 
its elliptical pupil in contrast to the 
rounded pupil of "'Dromicus." However, 
variation between ferox from the southwest 
population and ferox from other areas on 
Hispaniola with respect to pupil shape 
makes this character suspect for use on the 
generic level in this group until more is 
kmown concerning postmortem effects on 
this structure. Nevertheless, the other 
peculiarities of ferox, such as its enlarged 
teeth and unusual head shape, probably 
warrant retention of the currently recog- 
nized genus Hypsirliyncliiis. The consistent 
absence of a loreal scale in the populations 
from the southwest peninsula of Haiti, plus 
the suggestion of a distinction in pupil 



24 Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 



shape, justify the use of the subspccific 
name fewx scalaris for them. 

Dunn (1932) divided the remaining spe- 
cies of the present cuntherigerus assem- 
blage into two additional genera based on 
the number of apical pits on the dorsal 
body scales: Ahophis was distinguished as 
having two pits and Dromicus one pit. 

An examination of the relationships 
within the cantherigerus assemblage in the 
present study indicates that scale pit num- 
ber may or may not be consistent with 
other characters at the generic knel, and 
therefore cannot always be used to define 
major taxa. The genus Hypsirlujnclms has 
one pit and the six species of Uromacer 
have none. The remaining West Indian 
species have two pits side by side at the 
tip of the dorsal body scales, whereas on 
the mainland, chomissonis, ongiistilineatus, 
and tachymenoides have a single apical pit. 
The three Galapagos species have two pits. 
The distinction between Philodnias and 
Conopliis made by Boulenger (1896) was 
based on the presence of one apical pit in 
the former and none in the latter. Osteo- 
logically these two genera are very close 
but, since a detailed analysis of these 
groups was not made as part of the present 
study, they are here considered pro\'ision- 
ally distinct genera. The geographic dis- 
tribution of scale pits and their suggested 
phyletic relationships are shown in Figure 
17, along with several other characters. 
From the evidence one might reasonably 
infer a primitive condition of two scale 
pits with reduction as indicated in die 
figure. 

Scale pits have served, in the past, to 
diagnose groups based on this f(>aturc> 
alone. However, it is clear from other 
characters that the use of scale pits to 
define major groups may result in over- 
splitting of otherwise closely related as- 
semblages. With the recognition that scale 
pits by themselves are useful as taxonomic 
characters within this group only at the 
species or species-group level, the West 
Indian species of the cantherigerus as- 



semblage (excluding Ilypsirhynchtis and 
Uromacer) may be considered congeneric 
with chamissonis, angiistilineatus, and 
tachymenoides from South America, and 
with dorsalis, hiserialis, and slevini on the 
Galapagos archipelago. 

Smith and Grant (1958) have shown that 
Bibron's (1843) type of Dromicus was 
cursor from Martinique. This is a form 
unrelated to South American "Dromicus," 
as I will show below. WUh the name 
Dromicus thus unavailable, the present as- 
semblage of species is referred to Adsophis 
with the type antillensis ( = leucomelas) 
Fitzinger. (See Brongersma, already cited 
above, for the identity of the name antillen- 
sis.) 

The suggested phyletic relationships be- 
tween the genera and species of the can- 
therigerus assemblage are shown in Figure 
18. 

Zoogeography. Two lines of evidence 
indicate a western origin for thc^ cantheri- 
gerus species assemblage into the West 
Indies. First, Alsophis cantherigerus from 
Cuba is the Antillean species most similar 
to A. chamissonis of the mainland; this 
similarity is most notable in osteological 
characters and especially in the dental 
formula, which in these two species is the 
lowest of the whole assemblage. The Gala- 
pagos forms, likewise, have relatively few 
teeth, as do the suggested mainland deriv- 
atives PJiilodryas and Conopliis. Within 
the West Indies a general trend toward 
increased number of teeth is evident, espe- 
cially in the specialized arboreal species of 
Uromacer, and in the portoricensis species 
group. From the geographic distribution 
of dental formulae, it would appear that 
a low number of teeth is primitixe for the 
mainland ancestor of this assemblage. In 
other characters also, the more easterly 
distributed species show a greater diver- 
gence from the mainland forms, thus tend- 
ing to support the \'iew that the group 
entered from the ^\'est. 

The second line of evidence indicating 
a western origin is that no member of this 



West Indian Xenodontine Colubrid Snakes • Maglio 25 



CO 



.CO 



^ ^ 05 
"co "^ -Q 



CO 

O O 
■"::: c: 



to 5q 

C o <l^ 

o ^ c: 

CO CD -^ 

.CO c; ->^ 

-XT ^ C 

O -S CD 



to 



tu 



S2 






.<o 




:3 




to 


to .^ 




sirhynd 
macer 


CO 


:3 
c: 


foricen 

cticruc, 
ersmai 


vent r is 
'Hen sis 


5l o 


O 


qj 


§g:g 

cx (o V, 


^"^ 


^^ 




5 






A Anguilla Bank 

e Great Bahama Bank 

c Cuba 

G Guadeloupe, Dominica 

GA Galapagos 

H Hispaniola 

Jamaica 

Nevis, St. Kitts, etc 

Puerto Rico 

St. Croix 
sA South America 
— oversea colonization 



Fig. 18. Suggested phyletic relationships between species of the cantherigerus species assemblage and related genera. 
Short horizontal lines indicate proposed oversea colonizations. Geographic distributions as indicated by lettered symbols. 



species asseml)lage occurs on the Lesser 
Antilles south of Dominica. This in itself 
is not significant since, as Gorman and 
Atkins ( 1969 ) have shown for Anolis, 
colonization does not necessarily proceed 
sequentially island by island along the 
chain of the Lesser Antilles. Nevertheless, 
taken with the first line of evidence, this 
assumes more importance. 

The following zoogeographical histoiy 



for the genus AIsoplus may be suggested. 
The assemblage appears to have been de- 
rived from an ancestral species probably 
not unlike Alsophis canflwrigenis in its 
osteological, hemipenial, and external mor- 
phology. From this widely distributed 
ancestral group in South America (and 
probably Central America as well), a single 
trans-Caribbean colonization could have 
resulted in the establishment of this group 



26 Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 



A.vudii 



J 



A.cantlierigerus 






SCALE 

100 200 

I I I 

MILES 





A.rijgersmai 
^portoriceQs^''"^""""' 

A.sancticrucisM \n 
A.ruflventri%^. 



0. 



Fig. 19. Proposed routes of colonization of the West Indies by the cantherigerus species assemblage. The arrows are not 
intended to represent exact paths. The distribution of extant species and genera is as indicated. It is not certain whether 
this group entered from Central or South America. 



on Cuba. This was followed by a sub- 
sequent dispersion and radiation to Jamaica 
(ater), the Little and Great Bahama banks 
(vii(Ui), and Hispaniola (melanichnus) (Fig. 
19). From Hispaniola an early stage of 
melanichnus gave rise to portoricensis on 
Puerto Rico. The four speeies of the Virgin 
Islands and Lesser Antilles, sancticrucis, 
lufiventris, rijp,er.smai, and antillensis, ap- 
pear to be part of a relatively reeent radi- 
ation of portoricensis, with differentiation 
on these geographically isolated islands. 
The remaining history of the group involves 
a series of inter-island colonizations to 
centrally placed Hispaniola from the more 
peripheral islands of the Greater Antilles. 
A second migration from Cuba by an early 
stage of cantheri^ienis seems best to ac- 
count for anomalus, \\'hich shows a closer 
affinity to the Cuban species than to 
melanichnus (Hispaniola). Alternatively, a 
back colonziation from Puerto Rico may 



have resulted in anomalus. The rather 
peculiar Hypsirhynchiis ferox is close to 
ater (Jamaica) with respect to its skull 
and hcnnipenis and possibly represents a 
Jamaica-Hispaniola migration, with sub- 
sequent specialization of Hypsirhynchtis 
resulting in its external differences. Another 
endemic genus on Hispaniola, Uromacer, 
may have been derived from an early form 
of H. ferox before the latter achieved its 
peculiar specializations. 

Because of its central location, Hispan- 
iola has received a greater number of 
species than any of the other islands, and 
it may have been competition between 
these species that led to the specializations 
of some of them. The giant form anomalus, 
the arboreal species of the genus Uromacer, 
and Hypsirhynchus may have differenti- 
ated as a means of dixiding up the habi- 
tat more efficiently. Competition among 
closely related and overlapping species 



West Indian Xenodontine Colubrid Snakes • Maglio 27 



MAXILLARY TEETH 

t\JOJCJCJOJ(\Jf\JCJCJ 

I I I \ I I I L_ 



I I I 



— melanotus n=3 

— perfuscus N=2 

— or not us H-Z 
— cursor u-2 



N=4 

n — \ — \ — \ — I — \ — I — I — \ — \ — I — \ — \ — r 

OJCJCVJCVJCkJCJCVJCVJCVJ 



juliae 



PTERYGOID TEETH 

O — ojro^io uJNajOvO — (NPO'tio 
cvjcvjrocvjcvjoj cjcvjcNjcOfOrotororoto 



J \ \ L 



J LJ 1_L 



melanotus 



perfuscus 

'ornatus 
— cursor 



juliae 



-| — I — \ — \ — I — I — \ — \ — r _ 

cvjojojcvjcvjcj cvJCMCMOjrOrorofOrorO 



n — I — \ — I — TT" 
O) O — CJ ro ^ 



PALATINE TEETH 

o — cvjio^io<r)f^oo<7> o — cvjro 
cvj rofororo 



J I L 



LJjll I I 



4V 



— melanotus 

— perfuscus 
. ornatus 
— cursor 



1 — r—\ — r 



I I im 



ih 



juliae 



_____ cvjrorOrOfO 



DENTARY TEETH 

(Dr-oocnO— cjfO^iouJh-oocno— cvi ro^ 
cvioj cJCvicvJCJCJcvJCNJCNJrorOfororo 



I I I I L_L 



I I I I 



_LJ L_L 



melanotus 



perfuscus 



.ornatus 



• cursor 



— juliae 



-1 — \ — [—] — I — I — m — I — n — I I I I \ n r 

____c>jcvjrvjcvj(\jojcvjc\jcvjcvjrotorof0f0 



Fig. 20. Observed ranges of variation in numbers of teetfi on each tooth-bearing element for the five West Indian species 
of the melanotus species assemblage. 



often results in the well-known phenom- 
enon of charaeter divergenee whereby 
the competmg forms become adapted ( first 
ecologically, then morphologically) to 
somewhat different aspects of the environ- 
ment, thus reducing competition. It is 
interesting to note that, even in the rela- 
tively less specialized Hispaniolan species 
anomalus and melonichnus, character di- 
vergence has progressed to a remarkable 
degree, so that these forms lie near the t\\ o 
extremes of variation for the entire assem- 
blage in many of their characters (see Figs. 
6, 8, 9, and 11). 

Another waif dispersal from the main- 
land to the Galapagos almost certainly 
resulted in the differentiation of Imerialis, 
(lorsalis, and sJcvini on these islands. Later, 



rear-fanged specialization within the an- 
cestral mainland genus pbssibly led to 
Philochyas and Conophis which displaced 
their antecedent from much of its former 
range. The three closely related species, 
chamissonis, taclujmenoides, and ongusti- 
Uneofus, have remained as relatively un- 
modified relicts of the original mainland 
stock, except for reduction in the number 
of scale pits. 

MELANOTUS SPECIES ASSEMBLAGE 

Included West Indian species: cursor 
Lacepede, Martinique; julioe (including 
marlae) Cope, Guadeloupe, Marie Galante, 
Dominica; melanotus Shaw, South America, 
Trinidad, ? Grenada; ornatus Gaeman, St. 
Lucia; po-juscus Cope, Barbados. 



28 Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 



o 

cvl 


o 


O 


o 

IT) 


O 

CD 


o 


o 

00 


o 

(7) 


o o 
o — 


CM 

1 


CJ 


CM 


CM 
1 


CJ 


CM 

1 


CM 


1 


ro ro 

1 1 














1 


mPlnnniu^ m = o 












N = 


2 _ 


1 


perfuscus 










1 


_1_ 


ornatus 


N=2 
N = 2 






1 




jullae ^ 


1 = 4 




— I— 
o 


1 

O 

ro 


, .J ... 

O 
1- 


o 

IT) 


1 
o 

CC 


— r- 

o 

r- 


— I— 
o 

CO 


— |— 
o 


1 1 

o o 
o — 



Fig. 21. Skull length width incJices for the five West 
Indiarn species of the melanotus species assemblage. Widths 
are taken at the otic region of the skull. 



Osteologxj. This assemblage of species is 
distributed from the island of Guadeloupe 
south to Trinidad. It is distinguished from 
the Alsophis cantherigerus species assem- 
blage and other Antillean xenodontines 
primarily by the shape of the prefrontal 
bone and b\' the structure of the hemipenis 
(discussed below). The prefrontal is long 
dorso^'entrally and narrow anteroposteriorly 
with a sharply pointed anterior projection 
at about midlength (see Fig. 5). 

The species for which this group is 
named, melanotus, occurs on Trinidad and 
Tobago and has been questionably re- 
ported from Grenada. It also occurs in 
Venezuela and Colombia. Like all mem- 
bers of this group melanotus is about half 
the size of A. cantherigerus and contrasts 
with it in the following skull characters: 
the postorbital region is proportionately 
longer; the frontal bone is relatively shorter 
anteroposteriorly with very little emargin- 
ation above the orbits; the skull is con- 
siderably more flattened dorsoventrally as 
compared with its width; as a result of the 
latter character, the interorbital partition 
formed by the parasphenoid bone is shorter 
in its dorsoventral extension; long, thin 
lateral processes extend back from the 
premaxilla in contrast to the short, ^^ide 
based processes of A. cantherigerus. The 
dental formula for melanotus is about 15 + 
2 maxillary, 10 palatine, 24 pterygoid, and 
19 dentary teeth (Fig. 20 for variation). 



The skull of the endemic species per- 
fuscus on the island of Barbados is pro- 
portionately longer and narrower than that 
of melanotus (Fig. 21). The supratemporal 
and quadrate bones are more elongated 
and comparatively narrower. The orbit is 
small. The postorbital bone lies far for- 
ward on the parietal and is nearly in con- 
tact with the frontal. In this last character 
perfuscus is distinct from other members 
of the present group; in the latter, a wide 
expansion of the parietal bone separates 
the postorbital from the frontal. In general 
skull proportions and in its low dental 
formula, perfuscus is closer to melanotus 
and other mainland representatixes of this 
assemblage than it is to other Antillean 
species on adjacent islands. 

The species ornatus from St. Lucia is 
slightly larger than melanotus (Trinidad) 
and has a somewhat broader skull. The 
prefrontal bone is of the melanotus type, 
but is \\'ider than in that species. In most 
other skull characters the two forms are 
very similar except for the dental formula, 
which is higher in ornatus. Clearly related 
to the latter is the species cursor from 
Martinique (Fig. 22), which differs only 
in the smaller size of the nasal bone. On 
Guadeloupe, Dominica, and Marie Galante 
the species juliae has a higher dental 
formula than cursor, and the supratemporal, 
quadrate, and premaxilla are greatly re- 
duced in relative size. A similar reduction 
is seen in some related mainland species 
such as bimaculatus, but this probably 
represents parallelism rather than an inde- 
pendent derivation from one of these main- 
land forms. 

It is suiprising that no member of this 
group has been reported from St. Vincent. 
This island forms an important intermediate 
stepping stone between Grenada and St. 
Lucia. The mongoose is widespread on this 
island, and it is possible that a formerly 
present species, endemic or not, has be- 
come extinct. 

External morphology. Externally the 
melanotus species assemblage is a homo- 



West Indian Xenodontine Colubrid Snakes • Maglio 29 






5 m m 



J I 1 — \ — I 



Fig. 22. Skull structure of Dromi'cus cursor, MCZ 6011a reversed). A, dorsal view; B, lateral vievv; C, ventral view. 



30 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 




Fig. 23. Hemipenis morphology in Dromi'cus cursor (MCZ 
6011) showing the typical apical awn of the melanofus 
species assemblage; semidiagrammatic. The organ is un- 
everted and dissected in situ. Approx. X 5. 



geneous one and in many respects is similar 
to the cantherigerus assemblage. As in the 
latter group there are eight supralabials, 
but here only the fourth and fifth enter 
the orbit. The number of scale rows may 
be 17 or 19, and the ventral scales number 
appro-ximately between 150 and 200. In 
contrast to Alsophis, the number of sub- 
caudals is rarely over 100. The anal plate 
is divided. All of these snakes are moder- 
ate in size and are roughly between 600 
and 1000 millimeters in total body length. 
A single apical pit is usually present on the 
dorsal body scales, but may be absent as in 
jiiliae mariae. 



Hemipenis. The hemipenis of cursor 
( Fig. 23 ) is shorter proportionately than in 
Alsophis cantherigerus, and the sulcus 
spermaticus is less deeply divided. As in 
that species, several rows of stout spines 
extend along the sides of the organ but, in 
contrast to it, small spines are also present 
between the diverging branches of the 
sulcus. The organ is generally weakly 
bifurcated and the apical ornament is dis- 
tinct from all other West Indian xenodon- 
tines. Here papillae are never present; 
instead, a series of membranous folds radi- 
ate from the apex and terminate in a 
transverse fold of tissue which encircles 
the tip of each lobe. This fold forms a 
well-defined apical disk on each lobe when 
the organ is everted. The sulcus forks and 
proceeds onto the disk and to the tip of 
each lobe. 

The structure of the hemipenis in the 
other species in this assemblage is essen- 
tially like that of cursor. 

Origin and Zoogeography. The melanotus 
group offers no problem of origin. This 
well-defined and closely related assem- 
blage is moiphologically continuous with 
the widespread series of species currently 
referred to the genus Leirnadophis, common 
on the mainland.^ In both its osteology and 
hemipenial morphology the type species of 
Leimadophis — L. almadensis — is clearly al- 
lied to the present group, and there are 
no external characters which would pre- 
clude such a relationship. It seems probable 
that a northward colonization of the Lesser 



^ Dromicus anuizouicus is known only from the 
type specimen, MCZ 2820, and may also be re- 
ferred to the present group. Its prefrontal bone 
is essentially of the melanotus type and the hemi- 
penis has a weak apical disk. In all its osteological 
characters this specimen may be distinguished 
from other members of the present group only in 
ha\ing those features, such as broad skull, tliin 
cranial bones, low, rounded crests, etc., which 
normally characterize juvenile specimens. Even 
the relatively weak disk on the hemipenis probably 
reflects the youthful condition of the specimen. 
It is thus quite possible that this form represents 
an immatmc specimen of a previously described 
South American species of the present assemblage. 



West Indian Xenodontine Colubrid Snakes • Maglio 



31 




SCALE 

100 200 



MILES 



D. perfuscus 



Fig. 24. Proposed routes of colonization of the West Indies by the melanotus species assemblage. Distribution of 
extant species is as indicated. 



Antilles by a mainland species similar to 
melanotus occurred in relatively recent 
times. From this species or its ancestor, 
ornatus (St. Lucia) was almost certainly 
derived, possibly via a now extinct inter- 
mediate species on St. Vincent. Successive 
overseas migration (Fig. 24) probably re- 
sulted in cursor on Martinique and jtilioe 
on Guadeloupe, Dominica, and Marie 
Galante. The Barbados species, perfuscus, 
appears to have been derived from one of 
the islands to the west, perhaps from St. 
Vincent or Grenada. 

Taxonomy. As mentioned above. Smith 
and Grant (1958) have sho\\'n that Bibron's 
(1843) type of Dromicus was cursor. With 
cursor and ahnadensis here considered as 
congeneric, Dromicus Bibron 1843 and 
Leimadophis Fitzinger 1843 become syno- 
nymic names for this assemblage. Although 
the actual dates of publication of these two 
names remain in question, December 31, 
1843, is now to be taken as the official 
publication date of Fitzinger's Sijstema 



Reptilium (Smith and Grant, 1958), and 
Bibron's Dromicus thus becomes the senior 
synonym for the present assemblage with 
Dromicus cursor as the type species.^ 

The genus Dromicus is vei-v similar to 
the South American genus Liophis Wagler 
1830. In all of the characters studied, 
Wagler's type of Liophis — L. cobella — is 
close to the present assemblage. The pre- 
frontal is like that of D. melanotus, the 
frontal is short with very little emargination 
above the orbits, and the interorbital par- 
tition is veiy small. The hemipenis of 
Liophis has a pair of well-developed apical 
disks as in D. melanotus, but differs in the 
presence of basal hooks (Roze, 1964). Body 



^ After the present paper was in manuscript 
form, it was brouglit to my attention that Drs. 
Albert Sehwartz and Richard Thomas reached the 
conchision of "Dromicus" cf. cursor - Leimadophis 
synonymy independently of the present autlior. 
Their conclusion was reported in a letter to Dr. 
E. E. Williams. 



32 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 



.O 

Ci 



CO 


CO 








^ 


:3 








o 


to 

:3 


CO 


o 


CD 


C3 


s^ 


o 


CO 


.D 


qS 

E 




C: 


:3 


^ 


Q. 


o 


o 






B Barbados 

D Dominico 

L Lesser Antilles 

M Martinique 

s St Lucia 

SA South America 

— oversea colonization 

Fig. 25. Suggested phyletic relationships between species 
of the me/onotus species assemblage and the genus Liophis. 
Short horizontal lines indicate proposed oversea coloniza- 
tions. 

size and scale counts are within the range 
of Dromicus (present concept). 

The question arises as to the vaUdity of 
the generic distinctions between ^'Leima- 
dophis" (i.e., Dromicus) and Liophis made 
by Roze (1964: 535). As we have seen, 
the presence or absence of scale pits may 
not be as important a distinction as for- 
merly belie\'ed. Thus, the only major 
difference between these two genera is in 
the maxillaiy dentition in which "Leim- 
adophis" has a diastema with much en- 



larged posterior teeth, whereas Liophis 
lacks a diastema and the posterior teeth are 
little enlarged ( Roze, 1964 ) . I suspect that 
these two groups will be considered as 
congeneric when better known, but on 
present evidence I here treat them as valid 
genera. 

The inferred phyletic relationships be- 
tween Liophis and the West Indian species 
of Dromicus are presented in Figure 25. 

ANDREAE SPECIES ASSEMBLAGE 

Included West Indian species: amlreae 
Reinhaedt and LxJTKEN, Cuba; parvifrons 
Cope, Hispaniola. 

OsteolofS,y. Only two species of this 
assemblage are extant in the West Indies, 
andreae on Cuba and parvifrons on Hispan- 
iola and nearby islands. They are generally 
considerably smaller in size than Alsophis 
and are about the size of Dromicus mela- 
nofus. In cranial osteology amlreae and 
parvifrons show features characteristic of 
both Alsophis and Dromicus (Fig. 26). The 
frontal bones are very long and narrow 
with a deep emargination above the orbits 
as in Alsophis, but unlike the pro- 
portionately shorter and \\ ider frontals of 
Dromicus (see Fig. 3). On the contrary, 
in the structure of the prefrontal bone, the 
present assemblage is close to Dromicus 
with a long and narrow prefrontal bearing 
a shaiply rounded anterior process. This 
is (juite distinct from the relati\'ely wider 
prefrontal of Alsophis in \vhich the an- 
terior surface is broadly rounded and the 
bone is proportionately wider (see Fig. 5). 
In general skull proportions andreae and 
parvifrons are closer to Alsophis than to 
Dromicus. The parasphenoid bone is very 
narrow as in A. cantheriiierus and other 
westerly distributed species of Alsophis, as 
well as mainland forms, but unlike the 
rather broad shape of that bone in the 
A. portoricensis species group or in Dromi- 
cus. The parasphenoid partition between 
the orbits extends high abo\e the trabecular 
canals, as in most species of Alsophis except 



West Indian Xenodontine Colubrid Snakes • Maglio 33 






5mm 



I I I I L 



Fig. 76. Skull structure of Antillophis parvilrons nov. gen., MCZ 77228. A, dorsal view; B, lateral view; C, ventral view. 



34 Bulletin Museum of Comparaiwe Zoology, Vol. 141, No. 1 




Fig. 27. Hemipenis morphology in Aniillophis parvitrons 
nov. gen., MCZ 60064; semidiagrammatic. The organ is 
uneverted and dissected in situ. Approx. X 5. 



portoricensis and related species. The j)re- 
niaxillary bones in andreae and pawifrons 
lack the long lateral process as in Dromicus. 

The dental formnlae in these two species 
is comparable to that of both Alsopliis and 
Dromicus; that is, about 16 + 2 niaxillaiy, 
12 palatine, 26 pterygoid, and 21 dentary 
teeth in parvifrons, and 21 + 2 maxillary, 
16 palatine, 35 pterygoid, and 26 dentaiy 
teeth in andreae. 

Nonosteological characters. In external 
characters andreae and pawifrons are 
similar to Dromicus. The number of ven- 
tral scales is generally lower for these 
species than in Alsopliis and about the 



same as in Dromicus. The subcaudal scale 
number in parvifrons is higher than that of 
andreae and of species of Dromicus, while 
it is within the normal range for species of 
Aho})]}is. 

In the present assemblage, each dorsal 
body scale bears a single sensory pit in 
contrast to the two pits in all West Indian 
species of Alsophis, and in this respect is 
like most species of Dromicus. 

It is in the structure of the hemipenis, 
however, that andreae and parvifrons de- 
part radically from Dromicus. Here the 
organ has a deeply dix'ided sulcus sperm- 
aticus, bordered by a thick fold of spinose 
tissue. Several rows of stout spines parallel 
the sulcus, each branch of which terminates 
on a diskless apex bearing long filiform 
calyces (Fig. 27). There are no basal 
hooks, neither are there spines between 
the branches of the sulcus spermaticus as 
in Dromicus. This structure is basically 
like that of Alsopliis and is quite unlike 
that of Dromicus in \\'hich strong apical 
disks are present. 

Taxonomy and Origin. As noted above, 
the two species of the andreae group ex- 
liibit osteological features characteristic of 
l)oth Alsophis and Dromicus, though closer 
to the former. In most external characters 
they seem closer to Dromicus, but in the 
structure of the hemipenis they differ 
markedly from that group, being extremely 
close to Alsophis. Taken as a whole, this 
group cannot easily be referred to either 
genus. With regard to both skull and hemi- 
penial characters, these two species re- 
semble certain members of the South 
American genus Lygophis. The hemipenis 
in members of the latter genus, as currently 
recognized, is very heterogeneous. The 
organ may possess well-differentiat(>d apical 
disks as in L. lineatus (Fig. 2SA), the type 
species, and L. flavifrenatus, or it may be 
essentially of the Alsophis type, as in L. 
hoursieri (Fig. 2SB), in which the apical 
calyces are more spinulate along their 
margins than in the Alsophis type. Of these 
three species, Lygophis hoursieri (Ecuador 



West Indian Xenodontine Colubrid Snakes • Magjio 35 





Fig. 28. Hemipenis morphology in the South American genus Lygophis; semldiagrammatic. The organ Is uneverted and 
dissected in situ. A, Lygophis lineatus, MCZ 80994; B, Lygophis boursieri, MCZ 36948. Approx. X 5. 



and Colombia) is very close to andreae 
and parvifrons in both skull and hemipenial 
characters, as well as in external scale pat- 
tern. 

From this incomplete study of Lygophis., 
it seems possible that we may be dealing 
with a compound genus of distantly re- 
lated forms. The evidence seems to sug- 
gest that L. boursieri might be placed in a 
separate genus with andreae and parvi- 
frons as a specialized radiation, perhaps 
derived from mainland Alsophis; and that 
the type of Lygophis, L. lineatus, may need 
to be synonymized with Dromicus from 
which it differs primarily in th(> lack of 



scale pits and the presence of basal hooks 
on the hemipenis (Roze, 1964). However, 
we must await more detailed taxonomic 
work to clarify these issues.^ Nevertheless, 
it seems clear that generic distinctions be- 
tween Dromicus (present concept) and 



^ In a reappraisal of South American snakes 
related to LijgopJiis boursieri, Myers ( 1969 ) 
established the boursieri species group including 
Rhadinaea antioquien,sis, R. tiistriata, and L. 
boursieri. The group is placed in Lijgophis 
"... simply because boursieri already resides 
there." But the artificiality of the resulting genus 
is clearly pointed out, and the necessity of generic 
reassignment is affirmed. 



36 Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 



Lygopliis (see Roze, 1964) rest on slim 
evidence. 

Since on present generic concepts it is 
not possible to accommodate the species 
of the ondreae group in any existing genus, 
it seems best to propose a new generic 
name. Though close similarities exist be- 
tween andreae and parvifrons and at least 
part of what is now called Lij<i,opJus, the 
type species, as stated above, is certainly 
distinct from them on the generic level. 
The solution adopted here, though not 
completely satisfactory, seems the most 
desirable one until a complete revision of 
the mainland forms is undertaken. 

Antillopliis nov. gen. 

Type species: Dromicus parvifrons Cope 
1862. Proc. Acad. Nat. Sci., Philadelphia, 
1862, p. 79. 

Distribution of genus: Hispaniola, Gonave 
Island, Vache Island, Tortue Island, 
Cuba, Isle of Pines. 

Diagnosis: Small to medium-sized xeno- 
dontine snakes having generally long and 
narrow frontal bones as in Ahophis, but 
narrow prefrontal bones as in Dromicus; 
scale rows 17-19; 9-10 lower labials, 8 
upper labials with nos. 3, 4, and 5 border- 
ing the orbit as in Alsophis, but unlike 
Dromicus in which only nos. 3 and 4 
generally border the orbit; one sensory 
pit on each dorsal body scale as usual in 
Dromicus, but unlike the usual condition 
in West Indian Alsophis or in Lygophis; 
hemipenis like that in Alsopltis, lacking 
the apical disks of Dromicus and the 
disks and basal hooks of Lygophis. 

Zoop^eography. On present evidence it is 
not possible to determine the zoogeographic 
history of this assemblage, and it seems 
fruitless to indulge in speculation at this 
time. If, as suggested, andreoe and parvi- 
frons are derivatives of a mainland stock 
close to L. boursieri, then their entrance 
into the West Indies must have involved a 
colonization independent of that of A/- 
sophis. It is possible, however, that further 



evidence will suggest a West Indian origin 
for these two species from the genus 
Alsophis. 

FUNEREUS SPECIES ASSEMBLAGE 

Included West Indian species: callilae- 
mus CossE, Jamaica; dolichurum Werner, 
Cuba; funercus Cope, Jamaica; haetiana 
Cochran, Hispaniola; polylcpis Buden, 
Jamaica; exi'guus Cope, Puerto Rico, Virgin 
Islands; taeniatum Qij-NYnFXi, Cuba; vittatum 
Gundlach, Cuba. 

O,steolois.y. This group of eight species 
is very distinctive and is clearly separable 
from other West Indian xenodontines. They 
are all very small in size. The prefrontal 
bone is unlike that of Alsophis or Dromicus 
( Fig. 5 ) . It is moderately long dorso- 
ventrally, narrow and gently curving with 
nearly parallel anterior and posterior edges. 
The two frontal bones togeth(>r form a 
square plate above the orbits, unlike the 
condition in other West Indian groups. 
Ventrally the parasphenoid is proportion- 
ately wider along its entire length, 
especially posteriorly. The skull is pro- 
portionately flatter, and thus the inter- 
orbital partition formed by the dorsal 
extension of the parasphenoid does not 
extend above the trabecular grooves. 
Rather, the frontal bone on each side covers 
the entire lateral aspect of the para- 
sphenoid. The supratemporal and cjuad- 
rate are reduced in size, and the latter is 
flat and triangular in shape. 

On Jamaica there are three species of 
this group, funereus, polylcpis, and calli- 
lacmus. The specific distinction between 
funercus and polylcpis has recently been 
demonstrated by Buden (1966), but osteo- 
logically they are very similar. They have 
the highest dental formulae of the group 
(Fig. 29) with about 19 + 2 maxillary, 11 
palatine, 19 pter\'goid, and 24 dentaiy 
teeth in funercus, and 17 + 2 maxillaiy, 
11 palatine, 24 pterygoid, and 27 dentary 
teeth in polylepis. The parasphc>noid bone 
beneath the orbits is broad throughout its 



West Indian Xenodontine Colubrid Snakes • Maglio 37 



MAXILLARY TEETH 
,^oocnO-cM;o^|ncDf-ooo^o-cvj 



I I I I I 1 



I I I I I 



• funereus N=2 
• polylepis N - 2 
— colli laemus N-Z 

— haetiana N-3 
exiguus N=4 



vittafum N=7 



— V. landoi N-3 
• dolichurum N = l 
taenia turn N = 2 



1 I r 



1 — \ — \ — I — I — r 



(-ooo^2z£d:25:!e^^?^g^^ 



PTERYGOID TEETH 

-r.mOOJ'a-CDOOOCVJ'd-lOCD 
yJCDii OJCJOJCMCJ 

I I I I I I I I I I I I 

• funereus 

— polylepis 
. callilaemus 

haetiana 

—^^—— exiguus 
— vittatum 

• / landoi 
• dolichurum 
— taenia turn 



T1 — \ — I — r 



I I I I 



U)Q0r:!:^:z.__c\jCJOJcM(\J 



PALATINE TEETH 

iou)t^co<J^2zS^!2^l£ 

I I I I I I 

• funereus 
. polylepis 

• callilaemus 

. haetiana 
exiguus 



vittafum 



• I', landoi 
• dolichurum 
— taeniotum 



~i ^ ^ I r 
lo (O r^ oo o) 



"T~i — r~i — r-r 

O — CJ lO ^ lO 



DENTARY TEETH 

I I I I I I LJ \ L_ 



• funereus 
_ polylepis 
callilaemus 
— - haetiana 
— — exiguus 



— vittatum 
— I/, landoi 
• dolichurum 
• taeniatum 

—[ — I — I — I — \ — I — I — I — \ — r 

OCVJ^lDCOOcJ^tDOO 
X____CVICJCJCJCM 



Fig. 29. Observed ranges of variation in numbers of teeth on each tooth-bearing element for the eight species of the 
funereus species assemblage. 



length, being wider anteriorly, and bears 
a deep midventral groove along its entire 
length (Fig. 30). The postorbital bone is 
short and stout and is separated from the 
frontal only by a small spur of the parietal. 
The juxtastapedial fossa leading to the 
fenestra ovalis is variable in the degree of 
closure around the columella; it may be 
widely open and rounded, formed equally 
by the prootic and exoccipital bones as in 
most other West Indian xenodontines (Fig. 
31A), or it may be constricted dorso- 
ventrally by a ventral extension of the 
prootic portion of the fossa. 



Also on Jamaica is the species callilaemus 
which is clearly related to funereus. The 
premaxilla and nasal are more solid and 
compact. The supratemporal and quad- 
rate are further reduced, but this is prob- 
ably a result of the smaller size of this 
species. The juxtastapedial fossa tends to 
be even more constricted than in funereus, 
and it may be nearly divided into two 
separate openings (Fig. 31B). The colu- 
mella extends out of the posterior opening 
and is directed more posteriorly than later- 
ally. The parasphenoid is wider than in 
funererus, and the midventral groove is 



38 



BiilJrtin Museum of Comparative Zoology, Vol. 141, No. 1 






5mm 

Fig. 30. Skull structure in Arrbyton funereus, MCZ 44901. A, dorsal view; B, lateral view; C, ventral view. 



more conspicuous. The dental formula is 
lower with about 12 + 2 maxillary, 7 pala- 
tine, 16 pterygoid, and 18 dentaiy teeth. 

At least three species found on Cuba 
may be referred to this assemblage. Of 
these vittatiun is closest to caUilaemus. 
The dentition is not strikingly different; 
there are approximately 14 + 2 maxillary, 
12 palatine, 10 pterygoid, and 16 dcntary 



teeth. The parasphenoid is variable in 
width as is the degree of closure of the 
juxtastapedial fossa. The premaxilla and 
nasal are massive elements which approach 
the type found in some burrowing forms. 
The orbit is small, and the postorbital bone 
is slender and in contact with the frontal 
bone; there is no intervening parietal spur. 
The supratemporal and cjuadrate are even 



West Indian Xenodontine Colubrid Snakes • Maglio 



39 



more reduced than in the Jamaican form. 
The otic region is expanded so that it is 
the widest part of the brain case. The sub- 
species vittatum landoi is like vittatum 
viffatutn in skull structure, but the dentition 
is somewhat reduced; there are about 
11 + 2 maxillaiy, 9 palatine, 9 pterygoid, 
and 14 dentary teeth. 

In another Cuban species, dolichurum, 
the parasphenoid bone is even broader than 
in vittatum and the premaxilla is more 
robust. The supratemporal is reduced to 
a mere splint, and further reduction in the 
postorbital bone is evident. In its dental 
formula dolichurum shows a reduction 
over vittatum kimloi and has about 10 + 2 
maxillaiy, 7 palatine, 6 pteiygoid, and 12 
dentary teeth. 

It is in taeniatum (Cuba) that we find 
the greatest development of the trends ob- 
served in the series leading from funereus 
and callilaemus on Jamaica to vittatum and 
dolichurum on Cuba. The parasphenoid is 
very broad with a very wide but shallow 
midventral groove. The skull (Fig. 32) is 
proportionately longer than in dolichurum, 
and the orbit is relatively smaller (see Fig. 
33 ) . Reduction of the supratemporal is not 
quite so great as in that species. The nasal 
is an elongated solid bone, and the pre- 
maxilla is massive and in firm contact with 
the nasals. The dental formula is the lowest 
of the entire assemblage with approxi- 
mately 6 + 2 maxillaiy, 5 palatine, 8 pteiy- 
goid, and 10 dentary teeth. The constriction 
of the juxtastapedial fossa is complete; 
bony dorsal and ventral growths from the 
prootic and exoccipital bones close off the 
center of the fossa, leaving two openings 
to the outside. The columella extends 
nearly straight back\\'ards out of the pos- 
terior opening, and the newly formed an- 
terior opening lies close to the foramen for 
the mandibular nerve (Fig. 31C). 

The Hispaniolan species of this assem- 
blage is hactiana, a montane form somewhat 
differentiated externalh', but ostcologically 
close to the other members of the group. 
The shape of the frontal bone is the same 





Fig. 31. Variation in the structure of the juxtastapedial 
fossa and foramen ovalis in the funereus species assemblage. 
A, Arrhyton funereus, MCZ 13294; B, A. callilaemus, MCZ 
69078; C, A. taeniatum, MCZ 19874. Roman numerals, 
foramina for cranial nerves. Abbreviations: of, "anterior 
foramen," see text; fo, foramen ovalis. Other abbreviations 
OS in Fig. 2. Approx. X 12. 



as in funereus (Jamaica), and the pre- 
frontal and postorbital are similar. As in 
the latter, the parietal spur between the 
frontal and postorbital is \'er\' small. In 
the degree of reduction in the supra- 
temporal and quadrate, and in general skull 



40 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 






5mm 



Fig. 32. Skull structure of Arrhyfon taeniatum, MCZ 44901. A, dorsal view; B, lateral view; C, ventral view. 



proportions, Jiaetiaiia is similar to funereus. 
The orbits are reduced in size nearly as 
much as in taeniatwn. The otic region is 
expanded as in the Jamaican species and 
the juxtastapedial fossa varies from widely 
open to completely closed. In lioetiano 
may be seen the beginning of a trend in 
a direction opposite from that seen in the 
callilaemiis-taeniatum series. Here the para- 



sphenoid is narrower than in funereus, 
and the dental formula is unreduced, with 
about 17 + 2 maxillary, 11 palatine, 24 
pterygoid, and 22 dentary teeth. 

The Puerto Rican species, exiguus stahli, 
is considerably larger than haetiana. The 
skull is proportionately narrower and as a 
consequence the frontals are relatively 
longer. The orbits are small as in other 



West Indian Xenodontine Colubrid Snakes • Maglio 41 



o 
o 


o 


o 


o 
ro 


o 


o 
in 




o 


o 


o 

CD 


o 


O 

o 


o 


o 

OJ 


o 

ro 


O 


o 
in 


o 

U3 


1 


1 


1 


t 

1 


1 


1 




1 


1 


1 


1 


in 

1 


in 

1 


in 

1 


in 

1 


in 

1 


in 

1 


in 

1 




1 




funereus 


N = 


2 


























1 


polylepis 


N= 


2 



























callilaemus N=2 



exiguus N = 4 

~i 



y. landoi N = 3 



-I — haetiana N= 2 



vittatum N=7 
N=2 



taeniatum 



o o 
o - 



o 

CVJ 



o 

ro 



O 



O 

in 



o 



o 



o 

00 



o 

CD 



o o 
o — 

in m 



o 
CVJ 

in 



o 
to 

in 



o 
in 



o 
m 

in 



o 
CO 

in 



Fig. 33. Skull length/orbit length indices for seven of the eight species of the funereus species assemblage (Arrhyfon 
do/ichurum is not included). Horizontal line, observed range; vertical line, mean; open rectangle, ± one standard devia- 
tion from the mean; solid rectangle, 95 per cent confidence interval of the mean. 



members of the group, but not as small 
relatively as in lioetiana. The parasphenoid 
is as wide as in the Hispaniolan species. 
The dental formula is approximately 15 + 2 
maxillary, 9 palatine, 23 pterygoid, and 22 
dentary teeth, and is nearly the same as in 
haetiana. On the Virgin Islands the sub- 
species exiguus exiguus is very similar to 
exiguus staliU in osteological characters 
except for the lower dental formula ( about 
13 + 2 maxillary, 7 palatine, 17 pterygoid, 
and 19 dentary teeth), and the more re- 
duced supratemporal. The species exiguus 
is superficially similar to Dromicus juliae 
(Dominica): the latter is small with a 
wide skull, short frontals, and reduced 
supratemporals. However, several impor- 
tant differences suggest that exiguus is not 
closely related to Dromicus, but that it is 
allied with the funereus assemblage. In 
juliae the parasphenoid is narrow and the 
interorbital partition is very prominent, as 
in Drotnicus. The prefrontal is of the 
Dromicus type and is unlike that of 
exiguus. The orbit is large and the (quad- 
rate is relatively long. In all of these 
characters D. juliae contrasts with exiguus. 



On these grounds (and on external and 
hemipenial evidence), exiguus is placed in 
the funereus group close to liaetiana. 

External morphology. Externally the 
funereus species group is more diverse 
dian the three other West Indian groups 
discussed above. The usual number of 
supralabials is seven, but may be seven or 
eight in haetiana and six or seven in doli- 
churum. In exiguus the usual number is 
eight. The infralabials number eight in 
the Jamaican species (funereus, polylepis, 
and callilaemus), eight or nine in the 
Cuban forms (taeniatum, vittatum, and 
dolichurum), usually eight and more rarely 
nine in haetiana, and nine in exiguus. The 
number of scale rows is 17 in all of the 
Cuban species and 19 in the remainder of 
the group. Ventral scale number may vary 
from a median of about 118 in vittatum to 
about 1(S1 in taeniatum; caudal counts vary 
from a median of about 45 in haetiana to 
one of about 119 in dolichurum. The loreal 
is absent in haetiana and taeniatum, but 
the method of scale loss in the two species 
was different. In taeniatum the prefrontal 
scale has either extended downward later- 



42 Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 





Fig. 34. Hemipenis morphology in the funereus species assemblage; semidiagrammatic. The organ is uneverted and dis- 
sected in situ. A, Arrhyton iunereus, MCZ 13295; B, A. vittatum, MCZ 42505. Approx. X 5. 



ally to take the plaee of the loreal and thus 
lies between the nasal and the preoeular, 
or else the loreal has fused with the frontal, 
giving the same result. In Imetiana the 
nasal has either replaced the loreal by 
being extended backward to contact the 
preoeular, or has fused with the loreal. 

Of the diagnostic characters used by 
Cochran ( 1941 ) in her description of 
Dorlingtonia liaetiana, only two set this 
species apart from the other members of 
the present species assemblage. These are: 
tlie single anal plate in contrast to the 
divided plate of the other species; and the 
first pair of lower labials not meeting in the 



midvcntral line. In view of the scale 
variations already noted for this group, 
these characters are only doubtfully of 
generic value in this assemblage. 

With respect to scale pits, funereus and 
polylcpis haw two pits per dorsal body 
scale, calUlaemus has only one. The pit in 
the latter is asymmetrical and suggests its 
derivation from a two-pit condition by the 
loss of one pit. The Cuban species lack 
scale pits, as does haetiana. On Puerto' 
Rico exiguiis stahli has no pits on most of 
the body scales, but may have two pits on 
the scales of the dorsal side of the neck. 
The subspecies exiguus exiguus has no 



West Indian Xenodontine Colubrid Snakes • Maglio 43 



pits. In view of the numerous other char- 
acters indicating close relationship, scale 
pits do not appear to be reliable as a taxo- 
nomic character above the species level in 
this group of snakes. Rozc (1958) came 
to the same conclusion with reference to 
"Urotheca," and I have suggested the same 
for Alsophis, above. 

Hemipenis. As seems to be the case with 
other characters, the structure of the hemi- 
penis is considerably more variable here 
than in other West Indian groups. In 
fimereus and pohjiepis the organ is un- 
divided although the sulcus spermaticus is 
deeply forked (Fig. 34A). Small spines 
are situated along the sulcus from the base 
to somewhat beyond the point of branch- 
ing, whereupon lateral bands of spines 
encircle the organ. In funereiis; spinose 
folds of tissue border the sulcus for much 
of its length, and fine spinose p)apillae 
cover the apex. In call Hoe miis, the hemi- 
penis is weakly bifurcated and bears 
several rows of relatively large spines along 
the sides of the sulcus. These grade into 
fine spines basally. The apex of each lobe 
is covered with soft spinose calyces. The 
hemipenis in taeniatum and vittatum (Fig. 
34B) is of the caUihemus type, but the 
lateral spines extend further towards the 
apex; the latter is covered with calyces of 
fine papillae instead of soft spines. This 
distinction, however, is veiy slight and the 
two forms are essentially the same. In 
haetiana the organ is essentially like that of 
caUihemus, but as in the Cuban species 
the spines extend more distally, and the 
apical ornament consists of papillate 
calyces. The Puerto Rican and Virgin 
Island species exiguus has a more deeply 
bifurcated hemipenis than does haetiana, 
and the area of strongly reticulated apical 
papillae is sharply demarcated around its 
edge. 

Figure 35 summarizes the geographic 
distribution of some of the more important 
morphological characters. 

Oripn. The fimereus species assemblage 
cannot easily be derived from any other 



West Indian group. Turning to the main- 
land, the genus Rhodinaea^ shows a re- 
markable similarity to the fimereus group 
as a whole.- 

The external scale pattern in species of 
Rhadinaea shows a variation equal to that 
of the fimereus group. The supralabials 
are usually eight in number, except in R. 
flavilata and R. caUigaster in which there 
are seven. The number of infralabials may 
be eight, nine, or ten. The fifth infralabial 
is the largest of the series in the fimereus 
group, but in Rhadinaea the largest may be 
the fourth, fifth, or sixth scale. Consider- 
able variation is seen in the structure of 
the hemipenis also. In most species of 
Rhadinaea, such as R. flavilata and R. 
decorata, the organ is not bifurcated and 
the sulcus spermaticus is only veiy weakly 
divided near the apex. Very large lateral 
spines are arranged in several rows along 
the sides of the sulcus. The apex is capitate 
with spinulate calyces arranged in several 
thick folds. In R. caUigaster the sulcus is 
more deeply divided. The apex is only 
weakly capitate and the thick folds are 
lacking; the hemipenis is similar to that of 
vittatum (Cuba), with papillate calyces 
rather than spinulate ones. 

In its osteology Rhadinaea is basically 
like the fimereus group. The shape of the 
prefrontal bone is distinct from the An- 
tillean forms for the most part; the ventral 
half is greatly constricted anteroposteriorly. 
However, this is variable and in some spe- 
cies this bone approaches the condition 
found in the fimereus assemblage. The 
skull is short and the supratemporal bone 
is reduced. In some forms, such as R. 



^ This generic name is used here in the sense 
of Myers ( 1967 ) and is considered equivalent to 
Urotheca soi.sti Roze (1958). 

- RJiadinaea dunwiiUii, the type species, was 
unavailable for study, but from Bibron's (1843) 
figure and description it appears to be close to 
Central American species of Rhadinaea and unlike 
any Cuban species. jR. dumeiillii now appears 
not to he a Cuban form as originally described, 
l)ut almost certainly a mainland form (Roze, 
1958). 



44 Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 



CUBA 



A. vittatum 



-► A . taeniafum 




no scale pits 
(12 + 2, 10) 

hemipenis weakly 
bifurcate; papillate 
calyces . 



no scale pits 
loreol absent 
(6 + 2,5) 

hemipenis weakly 
bifurcate; papillate 
calyces . 



HISPANIOLA 

D- haetiana 

no scale pits 
loreal absent 
(16 + 2,11) 
hemipenis weakly 

PUERTO RICO bifurcate; papillate 

A- exiguus calyces. 

no scale pits (2) 
(13 + 2,7) 

hemipenis bifurcate; 
papillate calyces, 
capitate. 

Fig. 35. Geographic distribution of several morphological characters in the funereus species assemblage. The circle 
represents the supposed center of dispersion for this group. 



(lecorata, the parietal spur between the 
frontal and the postorbital i.s short as in 
fnnereus. In R. serperaster and R. lateri- 
stri<i,ci the parasphenoid bone is veiy wide 
and bears a deep niidxentral groove. All 
of the eharaeters that distinguish the 
fnnereus assemblage from other West 
Indian xenodontines also oceur within 
Rhadinaea, although no one of the speeies 
examined possesses all of these eharaeters. 
Although it seems likely that no one living 
species of Rhadinaea can no\\' be con- 
sidered as the ancestor of the Wcsi Indian 
funereus group, it is not unreasonable to 
suggest a common descent for both groups 
from an ancestral form combining the req- 
uisite characters, all of which occur in 



various combinations throughout this mor- 
phologically broad genus as presently de- 
fined. 

It is, of comse, possible that the moipho- 
logical similarities between Rhadinaea and 
the fnnereus group are the result of hal)itus 
rather than heritage characters, and that 
both groups have evolved in parallel 
fashion with respect to those characters 
related to a semiburrowing mode of life. 
There appear to be certain features com- 
mon to most semiburrowing xenodontines, 
and these almost certainly evolxt^d inde- 
pc>ndently in many groups vmder the in- 
fluence of similar habitus selection. But 
the exact way in which parallel characters 
are achieved, even under identical selection 



West Indian Xenodontine Colubrid Snakes • Maglio 45 



pressures, depends upon raw materials in 
the form of existing moqohological struc- 
tures, and upon genetic variability. The 
more distantly related any two forms are, 
the more likely it is that they will achieve 
functionally similar adaptations in a di- 
vergent way. 

An examination of semiburrowing adap- 
tations in xenodontine snakes of the New 
World shows similar osteological trends 
common to all of them, but, as expected, 
they differ from each other in details. In 
most osteological characters studied, Rha- 
dinaea and the funereus group exhibit a 
similarity of form which suggests more 
than morphological parallelism with re- 
spect to semiburrowing adaptations. 

The osteological modifications which 
generally appear to be associated with 
semiburrowing adaptations are: small body 
size; reduction in relative orbit size; short- 
ening of the parietal region of the skull; 
enlargement and consolidation of the pre- 
orbital bones to form firm contacts with 
each other; broadening of the parasphenoid 
bone associated with the reduced orbits; 
relative broadening of the otic region so 
that it becomes the widest part of the skull 
(probably associated with general stream- 
lining ) ; reduction of the supratcmporal and 
quadrate bones; and, a relatively low num- 
ber of teeth. 

With respect to all of these characters, 
as well as others not obviously correlated 
with burrowing, the funereus group and 
Rhodinoea show a close correspondence, 
differing only in the structure of the pre- 
frontal bone. Tlie frontal bones form a 
nearly square plate above the orbits and 
contact the parietals in a broad, nearly 
straight suture. In contrast, the parietals 
of Geophis miititorqids extend lateral to 
the posterior half of the frontals, occupying 
a deep groove in the latter. A similar con- 
dition exists in ApostoJepis am])imii,ra, 
where the frontals are trapezoidal in shape, 
being wider anteriorly. In Atractus Jati- 
frons the frontals are oval in shape, with 
their long axis in the transverse direction. 



In the funereus group and Rhadinaea 
the parietal retains the shape normal for 
nonburrowing xenodontines. In Geophis 
mutitorquis a posterior process extends into 
the deeply divided interparietal, and in 
Atractus lotifrons and Rhinostoma guia- 
nense the parietals are so shortened that 
they are broader than long. 

The reduced quadrate in the funereus 
group and Rhadinaea is thin and tri- 
angular in shape. In Rhinostoma <i,uianense 
and Drepanoides eatoni the quadrate re- 
tains its normal rodlike shape. It should be 
emphasized that in xenodontine snakes 
generally, the quadrate and supratcmporal 
appear to increase allometrically with body 
size and, therefore, the reduced size of 
these elements in most semiburrowing 
forms may, in part, be the result of their 
small size. 

The funereus group and Rhadinaea have 
retained more or less normal skull pro- 
portions, except for A. taeniatum in which 
the skull is somewhat elongated. In Rhino- 
stoma guianense the skull is much short- 
ened, whereas in ApostoJepis amhinigra it 
is greatly elongated. 

In the funereus group and in Rhadinaea 
only taeniatum shows some enlargement 
of the nasals and premaxilla and a definite 
trend towards consolidation of the pre- 
orbital region of the skull. In Carphophis 
amoena and ApostoJepis amlnnigra the 
nasals are greatly enlarged and form a firm 
contact with the frontals and premaxilla. 

In general skull details, Rlmdinaea and 
the funereus group approximate each other 
in nvnnerous ways and contrast with most 
other semiburro\\'ing xenodontines from the 
mainland, although they show a closer 
resemblance to forms like Drepanoides 
than to others like Rlunostoma, ApostoJepis, 
and Carphophis. The close similarity be- 
tween Rlmdinaea and the funereus group 
in osteological and other characters 
(excepting the hcmipenis) suggests a 
phylogenetic relationship rather than mor- 
phological convergence, though additional 
evidence is needed. The differences in the 



46 Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 



heniipenis between these t\\'o groups rests 
primarily in the eapitation of the organ in 
Rhadinaeo and the lack of eapitation in the 
funereus assemblage. I do not, however, 
feel that this difference is so significant as 
to preclude th(M'r bc^longing to a phylo- 
genetically related group of genera. 

It was stated above that vittatum landoi 
was distinct from vittatum vittatum in its 
dentition. In this respect v. landoi is inter- 
mediate between vittatum vittatum and 
dolichurum. Schwartz (1965) discussed 
the differences in ventral and subcaudal 
scale count and body length between the 
two subspecies of vittatum. These char- 
acters, together with the clear difference 
in dentition, indicate a distinction worthy 
of specific recognition, and the taxon landoi 
should probably be raised to the rank of 
full species. However, since the present 
sample of v. landoi was too small (three 
specimens) to determine the degree of vari- 
ation, it is here, for the present, retained 
as a subspecies of vittatum.^ 

Despite the morphological variation 
within the funereus group, it is difficult 
to divide these species into distinct genera. 
Although usually placed in different 
genera, vittatum (Cuba) is closer to calli- 
laemus (Jamaica) in dentition, osteology, 
and external scale pattern than it is to 
taeniatum. Both vittatum and callilaemus 
form intermediate grades between funereus 
and taeniatum, and a generic boundary 
within this group cannot be distinguished 
adequately. As discussed above, exiguus 
(Puerto Rico) shows certain similarities to 
species of the genus Dromicus. However, 
its totality of characters, especially the 
hemipenis, makes a close relationship be- 
tween the two unlikely. On present evi- 
dence exiguus seems allied to the present 
assemblage. The three species, fu7iereus, 
haetiana, and exiguus, form a moipho- 
logical series distinct from that leading to 



^ After the present paper was sent to press, 
Lando and Williams (1970) fonnally raised 
landoi to the rank of full species. 



taeniatum, although haetiana appears to 
have diverged from both funereus and 
exiguus in certain external characters. 
These characters have been used to sepa- 
rate haetiana from other West Indian spe- 
cies on the generic level (see above). It is 
my feeling that generic splitting within the 
present species assemblage will obscure the 
clear relationships between all of these 
species which ( except possibly for exiguus) 
certainly represent a phylogenetieally re- 
lated group. However, constancy in classi- 
fication makes the retention of the genus 
DarUngtonia for haetiana desirable at this 
time. 

The exact phyletic relationships between 
exiguus and other West Indian species is, 
as indicated above, not completely certain. 
Although open to question, I feel that this 
species could be considered congeneric 
with the funereus group without unduly 
broadening the limits of that assemblage. 

Although the morphological similarities 
between the genus Rhadirmea and the 
funereus assemblage possibly suggest an 
ancestor-descendant relationship (or per- 
haps a more distant common ancestiy), 
the funereus group is generically distinct 
from Rhadinaea on present evidence. The 
funereus assemblage, then, should be re- 
ferred to the genus Arrhtjton, which has 
priority, and the type species, unfortu- 
nately, is A. taeniatum, the most specialized 
form. The proposed phyletic relationships 
between Rhadinaea and the species of the 
genus Arrhijton are summarized in Figure 
36. 

Zoogeography. In terms of species 
diversity the center of distribution of 
Rhadinaea today is Central America. Be- 
ginning from a closely related stock, we 
may suggest the following zoogeographic 
history for tliis group. From an early 
stock of this (or an ancestral) genus a 
single oversea colonization presumably re- 
sulted in the establishment of the Arrhtjton 
funereus prototype on Jamaica. Arrhyton 
polylepis is extremely close to A. funereus 
and th(> two forms appear to be relatively 



West Indian Xenodontine Colubrid Snakes • Maglio 47 



C3 

O 

Q: 



to to 



03 



CD 



W) 














=3 






E 


S 

rx 




CJ 


CD 


o 

.Cj 




_2 


—J 




c: 


— *^ 


*•*«..» 


^ 


■ -«^ 


Qj 




03 


O 


o 




i» 


Ci 


>< 


O 


CJ 


T3 


ii 


:^ 




cb 


-c: 




c Cuba 

CA Central America 

H Hispaniola 

J Jamaica 

p Puerto Rico 

V Virgin Islands 

— oversea colonization 



Fig. 36. Suggested phylefic relationships between the species of the funereus species assemblage and the genus Rhad- 
inaea. Short horizontal lines indicate proposed oversea colonizations. 



recent products of speciation on that island. 
The most primitive member of the genus is 
A. funereus which forms the base of both 
the A. funereus-A. toeniatiim series and the 
A. funereus-A. exiguus series. It is the only 
Antillcan species (except for A. pohjiepis) 
with a nonbifurcated hemipenis and in this 
respect is closest to species of Rliadinaea. 
Within the A. funereus-A. taeniatum series 
it is the least specialized in terms of the 
reduction of skull bones and other semi- 
burrowing adaptations. From this early A. 
funereus stock two basic lines appear to 
have been established; A. coIUkiemus 
(Jamaica) began a trend toward reduction 
of the posterior skull bones and in the 
dentition, and a widening of the para- 



sphenoid bone. A colonization of Cuba 
from this early A. calliloemus stock led to 
the Cuban forms which, through repeated 
speciation, continued the trend, with A. 
taeniatum as the most highly specialized 
species of the group. The second line re- 
sulted from a migration to Hispaniola and 
tended toward a narrower parasphenoid, 
larger size, and in many ways a general 
convergence toward other W'cst Indian 
xenodontine groups, especially Dromicus. 
This may have been related to a general 
tendency away from semiburrowing adap- 
tations. In order to derive A. exiguus from 
this early Hispaniolan form we must postu- 
late a widespread distribution for this latter 
form before (or concurrent with) its differ- 



48 Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 






SCALE 

100 200 
I I 1 

MILES 




D. haetiana 



A . funereus ^ ■ exiguus ^^ 

A . poly le pis 








Fig. 37. Proposed routes of colonization of the West Indies by the funereus species assemblage. The arrows are not in- 
tended to represent exact paths. The distribution of extant species is as indicated. 



entiation into the morphologically some- 
what specialized and ecologically restricted 
species Darlingtonia haetiana. Today D. 
haetiana occurs only in the montane massifs 
of the southwestern and Barahona penin- 
sulas at altitudes ranging from 1000 to 5600 
feet. One possible explanation for the 
peculiar distribution of this species is an 
ecological replacement of the former wide- 
spread species (possibly by pawifrons?) 
with D. haetiana remaining as a montane 
relict. This zoogeographic pattern is sum- 
marized in Figure 37. 

A PROBLEMATICAL GENUS 

Two species remain to be discussed: 
laltris (lor.salis and 7. parlshi from Hispan- 
iola. laltris dorsalis (I have not seen I. 
pari.shi) Ls a large species and is most 
similar to Alsophis in many skull char- 
acters, but distinct in many external and 
hemipenial characters. The skull is nar- 
rower, especially in the otic region. In its 



dentition this species is unique among West 
Indian xenodontines in having prominent 
grooves on the enlarged posterior maxillary 
teeth. The bilobed hemipenis is very long, 
ridged with numerous folds, and has an 
apical ornament of weakly developed 
flounces. Externally there are seven up- 
per labials as in Arrhijton generally, but the 
\'entral and subcaudal scale counts are 
similar to those of Alsophis. laltris dorsalis 
(and presumably I. parishi also) is not 
very close to any other Antillean species 
as far as can be determined from its 
present morphological specializations and 
therefore certainly should be retained in 
a distinct genus. In most characters laltris 
shows its greatest similarity to Alsophis, 
and it may have been derived from that 
genus on Hispaniola. 

CONCLUSIONS 

The use of skull and hemipenial char- 
acters, in addition to those of external 



West Indian Xenodontine Colubrid Snakes • Maglio 49 



Table 2 
Distinguishing charactebistics of the eight genera of xenodontine colubrid snakes in the West 
Indies as discussed in the text. 



Genus 


prefrontal 


frontal 


hemipenis 


size 


supra- 
labials 


grooved 
max. 
teeth 


I 

anal 
scale 


no. 

sensor>- 

pits 


loreal 
scale 


Ahophis 


cantherigeriis 
type 


long and 
narrow 


no disk- 


large 


8 


no 


divided 


2(1) 


present 


Dromicus 


mehnotiis 
type 


short and 
narrow 


apical disk 


medium 


8 


no 


divided 


1(0) 


present 


AiitiJloiyJii.s- 


mdanotiis 
type 


long and 
narrow 


no disk 


medium 


8 


no 


divided 


1 


present 


Arrlnjton 


f tine reus 
t>pe 


sqnare 


no disk 


small 


7-8 


no 


divided 


0-1 


present 
(absent in 
A. taeniatum) 


Durlingionia 


fiiuereus 


square 


no disk 


small 


7-8 


no 


complete 





absent 



type 

Hijpsirhijnchus cantherigeriis long and no disk 

type narrow 

Uromacer cantherigerus long and no disk 

type narrow 



large 



no divided 1 present 



lalt 



ris 



cantherigerus long and no disk 
type narrow 



large, 8 no divided present 

arboreal 

large 7 yes dixided present 



to be of significant 



suggesting relationships between 



morphology, appears 
aid in 

species of West Indian xenodontine eohi- 
brid snakes. They not only provide data 
for a proposed redefinition of generic con- 
cepts, but suggest certain phylogenetic 
relationships with mainland groups. Such 
relationships are of considerable interest, 
since they allow a tentative reconstruction 
of the possible origin and history of these 
snakes in the Antilles. The generic groups 
of xenodontine snakes here recognized in 
the West Indies and listed in Table 1 may 
be distinguished as in Table 2. 

The present xenodontine fauna of the 
West Indies was possibly wholly derived 
from Central and South American stocks 
through at least four oversea colonizations. 
Based on present evidence, a summaiy of 
the postulated historical events follows: 
1) From the formerly widespread South 
American genus Alsophis, a waif coloni- 
zation established this group on Cuba. Sub- 



sequent radiation into a number of species 
and endemic genera led to its present 
distribution throughout the Greater An- 
tilles and the northern Lesser Antilles. A 
minimum of three separate inter-island 
migrations of this group is required to 
explain the peculiar faunal assemblage of 
Hispaniola. 2) The specialized genus laJtris 
possibly emerged from Alsophis on Hispan- 
iola. 3) Using Jamaica as a port of entry 
and center of dispersion, a single stock, 
possibly derived from the Central Ameri- 
can genus RhacUnaea, successfully spread 
through the Greater Antilles in two distinct 
but closely related lines and established the 
genera Arrliyton and Dorlingtonia. 4) Pos- 
sibly derived from part of what is now 
called Lijg,oph\s in Soutli America, the 
species andreae and parvifrons may have 
reached Cuba and Hispaniola by a direct 
oversea colonization. 5) A relatively recent 
invasion of the Lesser Antilles by a species 
of Dromicus ( = Leimadophis) almost cer- 



50 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 1 



tainh' (Mitered vhx Trinidad, but has not 
yet progrt^ssed be>'ond (iiiadcloupe. 

The chronological sequence of coloni- 
zations cannot definitely be established on 
present evidence. However, a sequence 
roughh' similar to that above is not un- 
reasonable. It, of course, cannot be as- 
sumed that the West Indies were devoid 
of a xenodontine ophi fauna before the 
series of colonizations that established the 
present fauna, but our knowledge of earlier 
xenodontine colonizers is nonexistent be- 
cause of the lack of a significant fossil 
record. 

From the patterns of dispersion discussed 
in this paper it would appear that numerous 
combinations of inter-island migrations 
have occurred. The main sequences have 
progressed from one island to the next 
adjacent island and in this sense were 
for the most part linear. The following 
series have been proposed: mainland-Cuba- 
Hispaniola-Puerto Rico-Lesser Antilles; 
mainland-Cuba-Jamaica-Hispaniola; main- 
land-Cuba-Bahamas; mainland-Jamaica- 
Cuba; mainland-Jamaica-Hispaniola-Puerto 
Rico; mainland-Trinidad-Lesser Antilles. 

Inter-island migration, especially to cen- 
trally located Hispaniola, seems to have 
been more frec^uent than mainland-island 
migrations. This was certainly the result of 
the greater cross-water distance between 
the mainland and any island than between 
the various islands themselves, as Simpson 
(1956) and Darhngton (1957) have sug- 
gested. The greatest diversity in species 
and genera occurs on Hispaniola; this is 
to be expected in view of its central position 
and consequently greater number of coloni- 
zations. Its large size, varied habitats, 
complex physiography and history have 
prox'ided an excellent opportunity for im- 
migrants to differentiate into noncompet- 
ing forms. 

The zoogeographical patterns here pro- 
posed are based on limited evidence and 
are in large measure speculative. It is 
hoped that they offer a workable contri- 
bution toward the continued study of this 



group. However, only when adequate in- 
formation about the comparative anatomy, 
karyotypes, ecology, physiology, and bio- 
chemistry of all Antillean snakes and their 
mainland relatives is available will we be 
able to draw firmer conclusions concerning 
the origin and zoogeography of xeno- 
dontine snakes in the West Indies. 

ACKNOWLEDGMENTS 

I wish to express my deep gratitude to 
Professor Ernest E. Williams for suggesting 
the problem which began this work, for 
innumerable discussions and suggestions, 
and for his constant encouragement. With- 
out his interest and assistance this study 
could never have been completed. 

For their reading of various versions of 
the manuscript and for many useful com- 
ments and criticisms, I extend my thanks to 
Drs. Richard Estes, George Gorman, Max 
Hecht, Edmond Malnate, Charles Myers, 
James Peters, Albert Schwartz, Richard 
Thomas, Paulo Vanzolini, and Ernest Wil- 
liams. The conclusions reached in this 
paper and any errors which remain are, of 
course, solely my responsibility. 

I am grateful to the following persons 
who kindly loaned specimens from their 
collections and ga\e permission to extract 
skulls: Dr. Richard Zweifel, Miss A. G. C. 
Grandison, Drs. Edmond Malnate, Albert 
Schwartz, and Ernest Williams. 

I wish thankfully to acknowledge Mr. 
Laszlo Meszoly \\'ho prepared the maps, 
graphs, and skull drawings, and Miss 
Catherine McGeary and Mrs. B. Gail 
Browne who typed various versions of the 
manuscript. The research was supported 
in part by National Science Foundation 
grants nos. GB-6944 and NSF B 019801X. 

LITERATURE CITED 

Barbour, T., and B. Shrevk. 1938. Concerning 
some Bahamian reptiles with some notes on' 
the fauna. Proc. Boston Soc. Nat. Hist., 40 
(5): 347-366. 

BiBROK, G. 1843. Reptiles, in Ramon De La 
Sagra, Histoire Physique PoHtique et Natur- 
elle de I'ile de Cuba, pp. 1-239. 



West Indian Xenodontine Colubrid Snakes • Maglio 51 



BouLENGER, G. A. 1894. Catalogue of the snakes 
in tlie British Museum (Natural History). 
London, vol. 2: 1-382. 

. 1896. Catalogue of the snakes in the 

British Nhiseum (Natural History). London, 
vol. 3: 1-727. 

Broxgersma, L. D. 1937. The types of Psam- 
moi)lii.s antillensis Schlegel, 1837. Heipeto- 
logical Note XIV, Zoologische Mededeelingen, 
20: 1-10. 

BuDEN, D. W. 1966. An evaluation of Jamaican 
Dromicus with the description of a new 
species. Breviora, No. 238: 1-10. 

Cochran, D. M. 1941. The Herpetology of 
Hispaniola. Bull. Smithson. Inst., U. S. Nat. 
Mus., No. 177: 1-398. 

Cope, E. D. 1862. Synopsis of the species of 
Holcosis and Ameiva, with diagnoses of new 
W. Indian and S. American Colnbridae. Proc. 
Acad. Nat. Sci., Philadelphia, 1863: 60-82. 

Darlington, P. J., Jr. 1938. The origin of the 
fauna of the Greater Antilles, with discussion 
of dispersal of animals over water and through 
the air. Quart. Rev. Biol., 13: 274-300. 

. 1957. Zoogeography. New York: John 

Wiley and Sons, 675 pp. 

Dowling, G. H., and J. M. Savage. 1960. A 
guide to the snake hemipenis: a survey of 
basic structure and systematic characters. 
Zoologica, 45 (1): 17-27. 

Dunn, E. R. 1928. A tentative key and arrange- 
ment of the American genera of Colnbridae. 
Bull. Antivenin Inst. Amer., 2(1): 18-24. 

. 1932. The colubrid snakes of the Greater 

Antilles. Copeia, 1932 (2): 92-98. 

Gorman, G. C, and L. Atkins. 1969. The zoo- 
geography of Lesser Antillean Atioli.s lizards- 
an analysis based upon chromosomes and 
lactic dehydrogenases. Bull. Mus. Comp. 
Zool., 138 (3): 53-80. 

Gosse, p. H. 1851. Naturalists' sojourn in Jamaica. 
London: Longman, Brown, Green, and Long- 
mans, pp. 1-508. 

Horn, H. 1969. Polymorphism and evolution of 
the Hispaniolan snake genus Uromaccr 
(Colnbridae). Breviora, No. 324: 1-23 

Lanix), R. v., and E. E. Williams. 1970. Notes 
on the herpetology of the U. S. Naval Base at 



Guantanamo Bay, Cuba. Studies of the fauna 
of Curacao and other Caribbean islands. No. 
116: 159-201. 

Mertens, E. 1939. Herpetologische Ergebnisse 
eine Reise nach der Insel Hispaniola, Westin- 
dien. Abh. Senckenberg. Naturforsch. Ges., 
No. 449: 1-84. 

Myers, C. W. 1967. The pine woods snake, 
Rhadinaca flacilata (Cope). Bull. Florida 
State Mus., 11 (2): 47-97. 

. 1969. South American snakes related to 

Lygophis hoiiisieii: a reappraisal of Rhadinaca 
antioquicnsis, Rhadinaca tristriata, Coronclla 
wJiymperi, and Liophis ataliiiallpae. Amer. 
Mus. Novitates, 2385: 1-27. 

RozE, J. A. 1958. A new species of the genus 
Urothcca (Serpentes: Colnbridae) from Vene- 
zuela. Breviora, No. 88: 1-5. 

. 1964. The snakes of the Leimadophis- 

Urotheca-Liophis complex from Parque Na- 
cional Henri Pittier ( Rancho Grande ) , Vene- 
zuela, with a description of a new genus and 
species ( Reptilia, Colnbridae ) . Senckenbergi- 
ana Biol., 45 (3/5): 533-542. 

Schwartz, A. 1965. Snakes of the genus Alsophis 
in Puerto Rico and the Virgin Islands. Studies 
of the fauna of Curacao and other Caribbean 
islands. No. 90: 177-227. 

. 1966. Review of the colubrid snake genus 

Arrhijton with a description of a new sub- 
species from southern Oriente Province, Cuba. 
Proc. Biol. Soc. Washington, 78: 99-114. 

Simpson, G. G. 1956. Zoogeography of West 
Indian land mammals. Amer. Mus. No\'itates, 
1759: 1-28. 

Smith, H. M., and C. Grant. 1958. The proper 
names for some Cuban snakes: an analysis of 
dates of publication of Ramon De La Sagra's 
Historia Natural de Cuba, and of Fitzinger's 
Systema Reptilium. Herpetologica, 14: 215- 
222. 

Underwood, G. 1962. Reptiles of the eastern 
Caribbean. Carib. Affairs (n.s.), Dept. Extra- 
mural studies, Univ. West Indies, No. 1 : 
1-92. 

. 1967. A contribution to the classification 

of snakes. British Mus. (Nat. Hist.), Pub. 
No. 653: 1-179. 



52 



BuUciin Museum of Comparative Zoology, Vol. 141, No. 1 



Appendix 
Dental pormxjlae and variation for West Indian and some related mainland and Galapagos 
xenodontine colubrid snakes. variation includes ranges of subspecies.'^ 





N2 


maxillary 


palatine 


pterygoid 


dentar>' 


Also phis 












angustilincaius 


1 


11 + 2 


11 


17 


18 


(iiunimlus 


1 


12+2 


8-9 


20 


18-19 


autillcn.'ii'i 


6 


16-19 + 2 


11-13 


28-30 


24-26 


ater 


4 


16+2 


i;3-16 


26-27 


22-25 


hi.scridlus 


1 


12 + 2 


7 


16 


17 


canihcri^erus 


12 


11-15 + 2 


9-12 


23-30 


17-21 


chammoni.1 


2 


9 + 2 


7-8 


10-14 


15-16 


(lor.saIi.s- 


1 


12+2 


7 


16 


17 


melauiclinus 


1 


18+2 


16 


28 


24 


partoricensis 


10 


14-18 + 2 


9-12 


26-32 


23-35 


rij^cr.smai 


2 


16-17+2 


11-13 


26-30 


23-24 


rufivcntris 


3 


16-18+2 


11-15 


25-28 


21-26 


sancticnicis 


2 


17-19+2 


12-13 


30-32 


22-23 


slevini 


1 


11 + 2 


8 


18 


19 


tachijmenoides 


1 


11 + 2 


10 


16 


21 


vudii vudii 


4 


11-13 + 2 


9-10 


20-27 


19-23 


viulii uiowanae 


2 


15 + 2 


12-13 


29-31 


22r-23 


Aniillophis gen. nov. 












andrcae 


4 


18-21 + 2 


15-16 


34-35 


25-28 


panifwns 


13 


13-16+2 


10-13 


25-27 


19-24 


Arrhyton 












callikiem i/.v 


2 


11-12 + 2 


7 


16 


18 


dolicliuium 


1 


10+2 


7 


16 


12 


exiguua 


4 


13-16+2 


7-11 


17-27 


19-23 


funcreus 


2 


19+2 


11 


19 


24 


])ohi\cpi.s 


2 


17+2 


11 


2.3-24 


26-27 


tiiciiiatum 


2 


6+2 


5 


7-8 


10 


vittatum vittatum 


7 


12-15+2 


10-14 


9-10 


15-17 


vitiattim Icmdoi 


3 


10-11+2 


9 


9 


14-15 


Conophis 












lineatus 


1 


10+2 


7 


17-18 


18 


Darlingtonia 












haetiana 


3 


16-17 + 2 


11 


22^25 


20-24 


Dromicus 












(dmadensi.s 


2 


18-19 + 2 


15-16 


27-28 


26-29 


amazonicus 


1 


18+2 


12 


22 


23 


hiniacidaiiis 


1 


21 + 2 


14 


28 


26 


cursor 


2 


20-21 + 2 


14-15 


27-28 


26 


epinc'phalus 


2 


22^24 + 2 


16-18 


27-32 


32 


jidiac 


4 


24-26 + 2 


29-33 


30-34 


33-34 


niclanohts 


3 


15-16 + 2 


10-11 


2.3-26 


16-22 


onuitus 


2 


19-20 


13 


26 


24-27 


})crfiiscus 


2 


15-16+2 


12-13 


20-22 


18-19 


pscudocobeUa 


1 


18+2 


13 


23 


19 


rcginae 


1 


16+2 


12 


27 


20 


taeniurus 


1 


19 + 2 


11 


25 


25 


Htjpsirhynchus 












ferox 


3 


11-12+2 


7 


17-19 


19-20 



West Indian Xenodontine Colubrid Snakes • Maglio 



53 



AprEXDix { Continued ) 



N2 



maxillarv 



palatine 



pterygoid 



dentary 



laliris 
dorsalis- 

Lio))Jiis- 

aiioDuila 
cohella 
jacgcri 
mcrrcmi 

Lijguphis 
hoursieri 
flavifrenutus 
liucatits 

Philodryas 

acsiivus 

Jntrmeisteri 

olfersii 

Uromaccr 
catcshtji 
dorsalis 
fremitus 
oxyrJiyucIiu.s 

Rhadinaca 
Incvirostrls 
d ecu rat (I 
flavdata 
■serperaster 



16 + 2 



21 



20 



1 


12 + 2 


10 


19 


17 


3 


19-20+2 


13-14 


26-29 


24-30 


1 


21 + 2 


18 


27 




1 


18+2 


14 


26 


22 


1 


22 + 2 


14 


28 


25 


1 


26+2 


23 


35 


35 


2 


19-20+2 


14-15 


29-32 


27-29 


1 


14+2 


11 


20 


19 


1 


12+2 


9 


16 


19 


1 


10+2 


9 


17 


13-14 


7 


15-18 + 2 


10-11 


24-25 


22-28 


2 


13-14+2 


9 


18 


20-22 


4 


14-16+2 


9 


18-19 


24 


7 


15-17 + 2 


8-10 


17-21 


22-28 


1 


14+2 


9 


21 


18 


1 


22+2 


14 


34- 


24 


1 


24+2 


11 


24 


20 


1 


17 + 2 


9 


14-16 


14 



1 Problematical subspecies which may be full species are listed separately. 
- Number of specimens examined. 






.vf-t 



OF THE 



Museum of 

Comparative 

Zoology 



The Milliped Family Conotylidae in North 

America, with A Description of the New 

Family Adritylidae (Diplopoda: Chordeumida) 



WILLIAM A. SHEAR 



HARVARD UNIVERSITY 

CAMBRIDGE, MASSACHUSETTS, U.S.A. 



VOLUME 141, NUMBER 2 
4 FEBRUARY 1971 



PUBLICATIONS ISSUED 

OR DISTRIBUTED BY THE 

MUSEUM OF COMPARATIVE ZOOLOGY 

HARVARD UNIVERSITY 

Bulletin 1863- 

Breviora 1952- 

Memoirs 1864-1938 

JoHNSONiA, Department of Mollusks, 1941- 

OccAsiONAL Papers on Mollusks, 1945- 



Other Publications. 

Bigelow, H. B., and W. C. Schroeder, 1953. Fishes of the Gulf of Maine. 
Reprint, $6.50 cloth. 

Brues, C. T., A. L. Melander, and F. M. Carpenter, 1954. Classification of 
Insects. $9.00 cloth. 

Creighton, W. S., 1950. The Ants of North America. Reprint, $10.00 cloth. 

Lyman, C. P., and A. R. Dawe (eds.), 1960. Symposium on Natural Mam- 
malian Hibernation. $3.00 paper, $4.50 cloth. 

Peters' Check-list of Birds of the World, vols. 2-7, 9, 10, 12-15. (Price list 
on request. ) 

Turner, R. D., 1966. A Survey and Illustrated Catalogue of the Teredinidae 
(Mollusca: Bivalvia). $8.00 cloth. 

Whittington, H. B., and W. D. I. Rolfe (eds.), 1963. Phylogeny and Evolution 
of Crustacea. $6.75 cloth. 

Proceedings of the New England Zoological Club 1899-1948. ( Complete sets 
only. ) 

Publications of the Boston Society of Natural History. 



Authors preparing manuscripts for the Bulletin of the Museum of Comparative 
Zoology or Breviora should send for the current Information and Instruction Sheet, 
available from Mrs. Penelope Lasnik, Editor, PubUcations Office, Museum of 
Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138, 
U.S.A. 



© The President and Fellows of Harvard College 1971. 



THE MILLIPED FAMILY CONOTYLIDAE IN NORTH AMERICA, 
WITH A DESCRIPTION OF THE NEW FAMILY ADRITYLIDAE 
(DIPLOPODA: CHORDEUMIDA) 



WILLIAM A. SHEAR 



ABSTRACT 

The milliped family Conotylidae in North 
America is revised; figures and descriptions of 
all known species are given. The new family 
Adritylidae is proposed for the genus Adrityla 
Causey. 

Two new genera of conotylids are described: 
Achemenides (type species, Conottjla pcctinata 
Causey) and Plumatyla (type species, Conotijla 
humerosa Loomis). Twelve new species of conoty- 
lids are described: Conotijla extonis, C. peisonata, 
C. elpenoi; C. smilax, C. octjpetes, C. aeto, C. 
vista, C. celcno, Taiyutijla napa, T. francisca, 
Aiistrotyla hoiealis, and A. chihuahua. Three new 
synonymies in the genus Conotijla are recognized. 

The Inology and zoogeography of the group is 
briefly discussed. 

INTRODUCTION 

The small niillipeds of the family Conoty- 
lidae are poorly represented in most col- 
lections, though species are fairly common 
in the northern parts of the United States, 
and at higher elevations elsewhere. That 
the distribution of this interesting group is 
a relict of a previous, colder age seems 
beyond question. Several species occur in 
caves, and many are characteristic of high 
altitudes, reaching near the timberline in 
the Rocky Mountains of Colorado and 
Alberta and in the White Mountains of 
New Hampshire, while adjacent lowlands 
are poor in species and in individuals. In 
the middle Appalachians, a j)attem of 
highly localized, rare, endemic species is 
now emerging. The importance of millipeds 



to zoogcographers was repeatedly empha- 
sized by O. F. Cook, but attempts to use 
them have been few, probably due to the 
chaotic state of diplopod taxonomy. It is 
hoped that studies like this one will en- 
courage zoogcographers and paleoecolo- 
gists to utilize the excellent zoogeographic 
information presented in many groups of 
millipeds. 

The family name Conotylidae was pro- 
posed by O. F. Cook in 1896 for the milli- 
ped genera Conotijla, Trichopetolum, Sco- 
feipes and Zij<;onopus, first described in 
detail by Cook and Collins in 1895. Since 
its establishment, little revisionary work 
has been done on the taxonomy of this 
family despite the large number of addi- 
tional names that have accumulated under 
it. Verhoeff (1932) realized that the 
Conotylidae, in the sense of previous work, 
was an unnatural assemblage, and removed 
some genera to his new family, Trichopeta- 
lidae; his paper was not noticed by 
American students. Chamberlin (1952) 
established the genus Taiyutyh, and 
several new species were added to Conotijla 
by Loomis ( 1939, 1943 ) and by Causey 
(1952). 

Hoffman ( 1961 ) , aware of Verhoeffs 
assignment of the genera Trichopetalum, 
Scoterpes, and Zijgonopus to the Tricho- 
petalidae, studied the gonopods of several 
species of Conotijla in detail. He reviewed 
previous studies of the conotylids, and 



Bull. Mus. Comp. Zool, 141(2) : 55-98, February, 1971 55 



56 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 2 



placed a number of the forms then in 
Conofijla into a new genus, Sonoratyla. 
Taiijutyla was also discussed, as well as 
Chaniberlin's enigmatic genera Zygotyh 
(1951), Cookella, and BoUmaneUa (1941). 
Hoffman made no attempt under the cir- 
cumstances (see section below on dubious 
names) to clarify the status of these latter 
three genera. At the same time, Causey 
(1961a) carried out a similar study, pub- 
lished earlier than Hoffman's, and proposed 
the genus Austrotyla for some of the spe- 
cies later included in Hoffman's Sonoratyla. 
Later in the same year, Causey (1961b) 
published the new genus, Adrityla, based 
on Conotyla deseretoe. Although further 
papers on conotylid genera have been an- 
nounced (Causey, 1961a), they have not 
as yet been published. 

T wish to thank Dr. Richard Hoffman, 
Radford, Virginia, for suggesting this study 
and providing numerous specimens and 
much unpublished data. His advice has 
been greatly appreciated. Dr. Nell Causey, 
Baton Rouge, Louisiana, loaned her col- 
lection of Conotylidae, probably the largest 
and most representative in existence, and 
made many helpful suggestions. H. F. 
Loomis, Miami, Florida, loaned unpub- 
lished drawings of new related taxa. Dr. 
Herbert W. Levi, Museum of Comparative 
Zoology, Cambridge, Massachusetts, read 
and edited the manuscript and loaned 
material from the collections under his 
care. Dr. Ralph Crabill, U. S. National 
Museum, loaned important type speci- 
mens. Dr. Andrew A. Weaver, Wooster, 
Ohio; Mr. Robin Leech, Edmonton, Al- 
berta, Canada; Mr. Michael Gardner, 
Davis, California; and, Mr. Erik Thorn of 
the British Columbia Proxincial Museum, 
Victoria, British Columbia, Canada, also 
loaned or donated important specimens. 
Mr. Stewart Peck of the Museum of Com- 
parative Zoology loaned specimens and 
unpublished data, and contributed much 
to my understanding of the biology and 
evolution of the cave forms in the Conoty- 
lidae. All types of new species described 



herein are deposited in the Museum of 
Comparative Zoology. 

TAXONOMIC POSITION OF THE 
FAMILY CONOTYLIDAE 

The affinities of the Conotylidae lie 
primarily with three North American fami- 
lies (one briefly described as new, below) 
and with a poorly known group of Asian 
and South American species. Verhoeff 
(1913) placed Japanosoma (Japan) in the 
monotypic subfamily Japanosominae, and 
Eudi^ona (Chile) in the subfamily Eudi- 
goninae of the Conotylidae. Neither of 
these subfamilies has been restudied since 
their original proposal, and their position 
remains in doubt. However, judging from 
the published data, they seem to be typical 
eonotylids. The Asian family Diplomarag- 
nidae is somewhat more distantly related 
to the Conotylidae (Verhoeff, 1942; Hoff- 
man, 1963). Recently, Buckett and Cardner 
(1967) have described a new monotypic 
family, Idagonidae, type genus Idop^ona, 
from caves in Idaho. They stated that the 
new family was related to both the Conoty- 
lidae and the Cleidogonidae, but as I have 
earlier pointed out (Shear, 1969), the 
Conotylidae and the Cleidogonidae are not 
at all closely related to each other. The 
single major distinction between the 
Conotylidae and the Idagonidae lies in the 
loss of the telopodite articles of the poste- 
rior gonopods (ninth legs) in the Idagoni- 
dae; except for this, Idagona might be 
considered as an aberrant conotylid genus. 
The relationship of the Conotylidae td 
the Trichopetalidae is somewhat prob- 
lematic. The two families are quite similar 
in genc^ral body plan, but there is a distinct 
morphological gap in gonopod structure. 
The anterior gonopods of the eonotylids arc 
fused into a single article; those of the 
trichopetalids are two-articled, with a 
prominent coxite. A colpocoxite is lackim: 
in the posterior gonopods of trichopetalids. 
which are two- or three-articled in all 
known forms; eonotylids have a large 



MiLLiPED Family Conotylidae • Shear 



57 



colpocoxite on the three-articled posterior 
gonopods. 

Milhped gonopods are developed from 
walking legs, and there is a well-recognized 
tendency for more primitive forms to have 
the gonopods more closely resembling legs; 
reduction and fusion of segments is con- 
sidered a specialized character. Thus, the 
Trichopetalidae seem a more primitive 
stock than the Conotylidae. Moreover, the 
conotylids have relatives in Asia and in 
South America, indicating possible disper- 
sal routes into North America from either 
of these regions (or the reverse), while 
the trichopetalids, like the clcidogonids, 
are peculiarly North American. My own 
feeling is that judgment should be reserved 
pending a study of the Trichopetalidae; it 
is quite likely that the resemblance be- 
tween them and the conotylids is due in 
great part to parallelism, or that tricho- 
petalids are nearer an ancestral stock than 
conotylids. The problem is aggravated by 
the obviously derived trichopetalid char- 
acters of reduction in segment number, 
loss of at least some ocelli, and minute size, 
all of which contrast with the seemingly 
more primitive gonopod plan. 

The following key will serve to separate 
the North American milliped families that 
may be confused with the Conotylidae. All 
of the families have in common swollen 
lateral segmental shoulders and greatly 
enlarged segmental setae; the male gono- 
pods superficially resemble each other.- 

Key to Conotylidae and Related Families 
IN North America 

la. Telopodites of ninth legs of males absent, 
nintli leg represented by the simple, sub- 
triangular colpocoxite, which curves around 
the larger telopodites of the anterior gono- 
pods; caves in Idaho 

Family IDAGONIDAE 

lb. Telopodites of ninth legs of males present 
but reduced, distal articles enlarged and 
turned dorsally 2 

2a. Tenth legs of males with the coxae greatly 
enlarged and lobed, tenth telopodite re- 
duced to two or three segments; anterior 
gonopods a pair of cheirites formed by 



fusion of telopodites, sternites, and tracheal 
apodemes Family ADRITYLIDAE, new 

2b. Tenth legs with the coxae only slightly 
enlarged, not lobed; anterior gonopods with 
telopodites free from the sternites 3 

3a. Ninth male legs with a conspicuous col- 
pocoxite, telopodites two-articled, the distal 
article oval and enlarged; anterior gonopods 
a single article; segmental setae usually less 

than one half the body width 

Family CONOTYLIDAE 

3b. Ninth male legs without colpocoxites, telop- 
odites usually a single article often witli 
conspicuous constrictions, often with a 
claw; telopodite of anterior gonopods two- 
articled; segmental setae frequently more 

than one half tlie Iwdy width 

Family TRICHOPETALIDAE 

A detailed study of the gonopod stmc- 
ture of Adrityla deseretoe (Chamberlin) 
has convinced me that it should not be 
retained in the Conotylidae, as to retain it 
would make the family obviously poly- 
phyletic. The new family Adritylidae, 
briefly characterized below, is probably 
related to the Conotylidae to about the 
same degree as is the Trichopetalidae. By 
even the most conservative criteria, the 
structure of the gonopods of Adrityla ex- 
cludes it from any known chordeumoid 
family, though there is a vague resemblance 
to Morquetia ( Opisthocheiridae; Brole- 
mann, 1935: 278); I have not seen speci- 
mens of Marquetia. 

Family ADRITYLIDAE new family 

Type genus. Adrityla Causey, 1961; type 
species Conotyla deseretae Chamberlin, by 
original designation and monotypy. 

Diagnosis. Large chordeumid millipeds 
(to 25 mm long) with 30 postcephalic 
segments, segmental setae and paranota 
prominent, head not covered by collum. 
Male gonopods modified from three inter- 
locking pairs of legs; anterior gonopods 
(eighth legs) consisting of pair of cheirites 
formed from fusion of telopodite, stemite 
and trachael apodeme of each side; poste- 
rior gonopods (ninth legs) three-seg- 
mented, coxae with large colpocoxite, 
IDartially fused to sternites and tracheal 



58 



BuUetin Museum of Comparative Zoology, Vol. 141, No. 2 



apodemos. distal telopodite article inflated, 
tumt^d dorsally. Tenth legs with coxae 
greatly enlarged and lobed, coxal gland 
present, coxae fused to sternites, telop- 
odites reduced. Eleventh legs without 
coxal glands. 

Notes. The singk^ known species, A. 
deseretae, is common in canyons of the 
Wasatch Mts., Utah. The reader is re- 
ferred to the paper of Causey (1961b) for 
further details. 

Family CONOTYUDAE Cook 

Conotylidae Cook, 1896, Biandtia, No. 2, p. 8. 
Veriioeff, 1932, Zool. Jahil). Abt. Syst., Vol. 62, 
p. 500. Hoffman, 1961, Trans. Amer. Ent. Soc, 
Vol. 87, p. 263. 

Type iienus. Conofi/Ja Cook and Collins, 
1895. 

Diagnosis. Small to medium-sized (9-25 
mm) chordeumid millipeds with 30 post- 
cephalic segments. Head not covered by 
collum. Eyes present, in triangular patch, 
sometimes reduced in size and number 
from maximum of 22-24. Antennae with 
third article longest, proportions of other 
articles variable; not markedly clavate; long 
and slender. Sensory cones four or five. 
Mouthparts typical. Mentum not divided, 
mandibles witli about 12 pectinate lamellae. 
Post-collum segments with prominent 
lateral swellings bearing outer two seg- 
mental setae on each side. Segmental setae 
large, prominent, movable. Epiproct and 
periprocts truncate; spinnerets two. Legs 
long, slender, claws prominent, basal seg- 
ments heavily setose-pilose. Legs one and 
two of males reduced in size, six-segmented; 
legs three through seven longer than post- 
gonopodal legs, usually thickened and eras- 
sate, often with strong knobs on some of the 
segments. Gonopods modified from eighth 
and ninth legs; anterior gonopods consisting 
of single article, sternum variable. Posterior 
gonopods three-jointed, coxa free from 
sternum, with large colpocoxite, distal telop- 
odite article inflated, turned dorsad, with- 
out a claw. Coxae of tenth legs with large 
coxal glands, coxae sometimes lobed. Coxae 



of eleventh legs without glands, sometimes 
with prominent hooks. Prefemur of eleventh 
legs with prominent hooks. Remaining legs 
unmodified. Female cyphopods of two 
valves opening anteriorly at rest, pore of 
oviduct covered by single receptacle. 

Distri])ution. North America, Japan, 
Chile. 

BIOLOGY OF THE CONOTYLIDAE 

Little is known of the biology and 
ecology of the Conotylidae. In southeast- 
ern West Virginia, immature specimens 
were frequently taken from leaf litter and 
rotted ^^'ood by Tullgren funnel sampling 
during the summer and early fall. Ho\\'- 
ever, nearly 90 percent of all examined 
collections of mature specimens, including 
those from caves and high altitudes, were 
taken between the months of November 
and April. Cook and Collins (1(S95) com- 
mented on the surprising activity of 
Conotyhi during the winter, and I have 
seen mature specimens of Conotyla walk- 
ing briskly over frozen logs and ice crystals. 
In addition, most of the records known to 
me from the eastern part of the continent 
are either from caves, high altitudes, or 
cool microclimates supporting such trees 
as hemlock, spruce, and fir. In the Rocky 
Mountains of Colorado, all but a few 
records of AiistrotyJa coloradensis are from 
coniferous forest abo\e 7000 ft. (2500 m) 
elevation. Conotyla alhcrtana has been col- 
lected in Alberta in alpine meadows above 
6500 ft. (2150 m), and no records below 
4200 ft. are known. Aiistrotyla specus, 
CoJiotyla blakei, and C. hollmuni are found 
at low elevations in the interior of the 
continent, but most reliable records are 
from caves. Achemenides pectinatus and 
Plumatyla humerosa are known only from 
caves, and the latter is the only conotylid 
showing well-marked cave adaptations. 

Howden (1963) has pointed out that 
animal species populations may adjust to a 
warming climate by persisting in cooler 
microhabitats (caves, mountain-tops), or 



MiLLiPED Family Conotylidae • Shear 59 



by adjusting their major period of activity vations on the ecology of only a single 

to the cooler part of the year. The evidence species, Conotyla hlakei. About 20 speci- 

abovc seems to indicate that, following mens of both sexes were observed and 

Quaternary glaciations, both adjustments collected on Mt. Equinox, Bennington Co., 

may have been made by various species Vermont, and Mt. Grey lock, Berkshire Co., 

of conotylids. Massachusetts. All specimens were taken 

The distributions of Conotyla hhkei from above 3000 ft. ( 1000 m ) in elevation, 
(Map 2) and of Austrotyla specus (Map in dense fir forests. The Mt. Equinox 
3) may reflect a current period of cave populations were observed in October, 
invasion and northward movement follow- 1968, and May, 1969; those on Mt. Grey- 
ing the Wisconsin glaciation and subse- lock in May, 1969. Without exception, the 
quent retreat. Barr ( 1968), in an excellent animals were found under the bark of dead 
summary of the evolution of troglobitic standing trees or fallen logs. The bark was 
animals, points out that the isolation and loosened to the extent that it could easily 
evolution of a troglobite usually is pre- be peeled off in large slabs, and the wood 
ceded by a stage during which an ances- underneath, on which the animals were 
tral species is troglophilic. The small cave usually resting, was saturated and soft- 
populations are occasionally swamped water could easily be squeezed from it by 
genetically by invasions from the sur- the fingers. Held in the hand or exposed 
face population. Isolation and subsequent to sunlight, the millipeds reacted by curl- 
speciation occurs when the surface popu- ing the head and first few segments under 
lations are eliminated by some (usually the body and withdrawing the legs into a 
chmatic) change that does not significantly position parallel with the long axis. After 
affect the environmental conditions in the a few seconds, a period of rapid running 
caves. Because populations in caves are and searching followed. If not allowed to 
small, genetic drift and founder phenomena find shelter from the sun, or if not released 
play a great part in their evolution (Barr, from the hand, the millipeds became im- 
1968: 82-84). Although present Conotyla mobile and moribund in a matter of 30 
and Austrotyla cave populations seem to be seconds. Similarly, an attempt to keep C. 
geographically isolated from large surface hlakei in the laboratory resulted in the 
populations, no appreciable differentiation death within 24 hours of all individuals 
has taken place. Perhaps more thorough collected, even though high humidity was 
surface collection in the regions of cave maintained. Presumably, temperature was 
records will produce evidence of popu- a major factor in the laboratory and field 
lations that can produce the swampmg deaths. 
^"^<^'*^' A single mating was obserx^ed in May, 

However, it must not be overlooked that 1969, on Mt. Equinox, at an elevation of 

it is the maxima and minima of climatic about 3800 ft.; there were numerous per- 

parameters, not the average, that usually sistent snowbanks in the fir thicket where 

restrict the range of a species. In the case the mating pair was found. The female 

of Achemenides pectmatus, caves may have held onto the wood of the rotted tree-trunk 

served as refugia from the severe peri- with the last 10 or 12 segments, while the 

glacial climate during the glacial maxima, male coiled about her body, using the cras- 

/V further difficulty is posed by the wide sate first seven pairs of legs to embrace 

distribution in California of Plumatyla her head and anterior segments. No thrust- 

humerosa, a true troglobite. More than one ing movements of any sort were obser\^ed, 

species may be represented in the nu- but the pair experienced difficult}' in 

mcrous collections of immature specimens, separating after being disturbed. 

I have made extensive personal obser- Schubart (1934) reported spermatophore 



60 



Bullet ill Mii.svuiii of Comparative Zoology, Vol. 141, No. 2 



formation and transfer in some European 
chordeumids. In these forms, the male se- 
cretes sperm from the seminal pores on the 
eoxae of the second legs into coxal sacs on 
the postgonopodal legs (in the case of the 
conot>'lids, these sacs are foimd on the 
tentli ](^gs only). The sc^eretions from these 
glandular sacs then form the seminal fluid 
into a spermatophore (Fig. 34) which, 
during mating, is transferred to the female 
cyphopods (Fig. 15) at the bases of the 
second legs by the gonopod complex (Fig. 
14) of the male. In presena^d male speci- 
mens of conotylids, these coxal sacs on the 
tenth legs are often extruded. In several 
cases, two large globular objects were at- 
tached to the extruded ends of the coxal 
sacs. These were easily detached, and 
examination under a compound microscope 
showed that each consisted of a cemented 
mass of small cells. This structure may 
represent a spermatophore. If this is the 
type of mating behavior that occurs in the 
family, then the distinction in gonopod 
function pointed out for other families by 
Brolemann (1935) and referred to by 
Hoffman ( 1961 ) may be applicable to 
conotylids.^ 

Feeding presumably takes place as in 
otlier chilognath millipeds, material being 
scraped or picked up from the substrate by 
a combination of movements of the gnatho- 
chilarium and mandibles, and ground by 
the dentate and pectinate lamellae of the 
mandibles. The finely triturated material 
in the foreguts of several C. hlokei was 
composed of wood tracheids and fungal 
hyphae. 

Molting and oviposition have not been 
obsei-ved, but some North American chord- 
eumids make silk chambers for these pur- 
poses. 



^ Because of some uncertainty, I have not used 
the terms paragonopod and peltogonopod (see 
Hoffman, 1961) but have used the more imder- 
standable, tiiough less specific, terms anterior 
gonopods to indicate those modified from the 
eighth legs, and posterior gonopods to indicate 
those modified from the ninth legs. 



TAXONOMIC CHARACTERS AND 
GONOPOD STRUCTURE 

Unfortunately, as is true of many milli- 
ped groups, nonsexual characters are of 
very little use in separating species, and 
even genera. Causey (1961a) attempted 
to use the relative lengths of the antennal 
segments in some species of Aiistrotylo, but 
such meristic characters must be used with 
caution when few specimens are available. 
In species that occur both in and out of 
caves, ocelli number and arrangement is 
variable, or it is the same in related epigean 
species, so that this character is usually of 
little value. Nonetheless, since such char- 
acters may at some time be useful, I have 
described them for all the species men- 
tioned here. I have, however, omitted 
much of the usual detail from the species 
descriptions, particularly description of 
structures which are common to all mem- 
bers of a family or genus. 

Also, female specimens are difficult to 
assign to species, and even genus. The 
cyphopods, useful in other chordeumid 
groups, are virtually identical in different 
species of Conotylidae. In addition, they 
are generally so poorly sclerotized that they 
shrivel when presei"ved, further reducing 
their utilit\^ in taxonomy. The best taxo- 
nomic characters are to be found in the 
male gonopods. In addition, the modi- 
fications of some of the pregonopodal legs 
of the males, usually taking the form of 
mesal lobes on one or two segments, are 
quite important in separating species. 

Conoti/hi melinda (Figs. 1, 2) has gono- 
pods typical in many respects of the large 
group of Appalachian species which are 
highly endemic in their distribution. In 
lateral view ( Fig. 1 ) the anterior gonopods 
are seen to curv^e laterad of the colpocoxites 
(c) of the posterior gonopods (telopodite 
articles of posterior gonopod not shown). 
The apex of the anterior gonopod is di- 
vided into two branches, an anterior branch 
(ah) and a posterior branch (ph), and there 
is a prominent lateral shoulder (Is) on 



MiLLiPED Family CoNOTiLiDAE • Shear 61 








■igures 1-6. Anatomy of conotylid male gonopods. Fig. 1. Conotyla melinda, lateral view. Fig. 2. C. melinda, anterior 
lew. Fig. 3. C. fiicheri, anterior view. Fig. 4. C. blakei, anterior view. Fig. 5. C. smilax, anterior view. Fig. 6. C. 
'trollneata, anterior view. 



62 Bulletin Museum of Comparative Zoology, Vol. 141, No. 2 



which, in some species, the process of the 
femur of the seventh leg rests (ConotyJa 
vista. Fig. 41). Note also the coxal sac of 
leg ten (cs) and the modified prefemur of 
leg eleven (pf 11). In an anterior semi- 
diagrammatic view (Fig. 2), the narrow, 
bandlike sternum of the anterior gonopods 
is visible, as well as the telopodite articles 
of the posterior gonopods. Hoffman (1961) 
showed that the narrow muscle (rcl) known 
as the remotor coxae longus, extending 
from the distal part of the tracheal apo- 
deme (ta) to the base of the anterior gono- 
pod, can be used to establish the homology 
between that portion and the coxa of a 
walking leg. Thus, the anterior gonopod is 
a coxotelopodite articulating directly with 
its sternum. In Conotijla fischeri (Fig. 3), 
a similar pattern is repeated, but the an- 
terior branch of the anterior gonopod (0^1,) 
is usually not developed. This branch is 
typically plumose when present. Conotijla 
hlakei (Fig. 4) is representative of a group 
of species with somewhat simpler gonopod 
construction. The anterior branch (ah) of 
the anterior gonopod is reduced to a poorly 
sclerotized plumose rod that is broken off 
in many specimens, and the posterior 
branch (ph) is a smooth continuation of 
the gonopod mass. There is no lateral 
shoulder. The colpocoxites (c) are likewise 
simpler in form (Fig. 17). The sterna are 
somewhat broader in this species than in 
the central Appalachian group. Conotijla 
similax (Fig. 5) is a slightly aberrant Ap- 
palachian form in which the anterior 
gonopods (ag) do not pass laterad of the 
colpocoxites (c), and bear no posterior 
branch. The colpocoxites (Fig. 31) are 
complex on the posterior surface, with a 
terminal process (tp), mesal tooth (mt), 
lateral notch (In), and pilose area (pa). 

Conotijla atrolineata (Fig. 6) differs 
from the eastern North American species 
of Conotijla in the simplification and re- 
duction of the anterior gonopods, the 
broadening of the sternum of the posterior 
gonopods (.s), and the much greater com- 
plexity of the colpocoxites (Figs. 48, 49, 



50). This trend culminates in the related 
species, C. alhertana, ^\'ith an exceptionally 
broad posterior gonopod sternum (Fig. 7). 

In Taiijiitijla corvallis (Fig. 8), the an- 
terior gonopod sternum (s) is heavily 
sclerotized and not bandlike, completely 
encircling the simple, platelike anterior 
gonopods (ag), which are larger than the 
colpocoxites (c). 

In Austrotyla colorademis (Fig. 9), the 
sternum of the anterior gonopods (.s) may, 
in certain preparations, appear divided, 
though it is in reality a single piece en- 
circling the gonopods and nearly meeting 
posteriorly. Lateral extensions of this 
sternum articulate with the lateral surface 
of the gonopod, which is simple and leaf- 
like and larger than the colpocoxite. The 
colpocoxite (c) is usually cupped anteriad 
and may have two parts. The tracheal 
apodemes are reduced in size. 

Plumatijla humerosa (Fig. 10) is a 
unique form somewhat intermediate be- 
tween Taiijutijla and Austrotyla, but the 
lateral extensions of the solidly constructed 
sternum (s) are more firmly fused to the 
gonopod than they are in either of the 
preceding. The anterior gonopod (ag) is 
complex (Fig. 80), with two branches per- 
haps homologous to the two branches fovmd 
in many Conotijla species. The colpocoxites 
(c) are remotely similar to those of Austro- 
tyla. 

Achemenides pectinatus (Fig. 11) dif- 
fers in most respects from the other mem- 
bers of the family. The anterior gonopod 
sternum (sj is truly divided, and each 
sternite is doubled over longitudinally 
(Figs. 54, 55). The anterior gonopods (ag) 
are fused at the base, and are much larger 
than the colpocoxites (c), which are widely 
separated on a thin, bandlike sternum. The 
tracheal apodemes of the posterior gono- 
pods are small and bifurcate. 

PROBLEMATICAL NAMES IN 
THE CONOTYLIDAE 

Due to the fact that the male genitalia 
are the only relialjle specific characters in 



MiLLiPED Family Conotylidae • Shear 



63 



this family, names based on females and on 
immature specimens are extremely difficult 
to assign, particularly if more than one 
species of conotylid occurs in the region. 
In two cases, Conotyki wyandotte (Boll- 
man) and Conotyh jonesi Chamberlin, I 
have made some attempt to place names 
attached to females or lost specimens in 
the proper synonymy. 

Listed below are the problematical 
names, \\'ith brief discussions of their 
original proposals and histories. 

Cookella leibergi (Cook and Collins) 

Conotyla leibergi Cook and Collins, 1895, T- New 
York Acad. Sci., 9(1): 77, figs. 102-104. 

Cookella leibergi, Chanilserlin. 1941, Bull Univ 
Utah, Biol. Ser., 6(5): 13. 

This species was originally described 
from a female (holotype in U. S. National 
Museum, examined), which was supposed 
to have a small, broadly triangular pro- 
mentum. This mistaken original observation 
of Cook and Collins (1895) was used by 
Chamberlin (1941) as a pretext for erect- 
ing the completely unnecessary genus 
Cookella. The type locahty, [the shore of] 
Lake Pend d'Oreille, Idaho, could be 
within the range of either Conotyla atro- 
lincata or C. albertana, and leibergi would 
thus be treated either as a junior synonvm 
of atroUneata or the correct name for al- 
bertana. Loomis (personal communication, 
196S) has in preparation descriptions of 
new taxa from the Idaho-Montana area; 
leibergi males may conceivably be repre- 
sented among them. As neither the generic 
or specific name can be properly dealt ^\•ith 
in the absence of males, I consider Cookella 
leibergi a nomen chibium. 

Trichopetalum glomeratum Harger 

Trichopetahim glomeratum Harger, 1872, Amer. 
^J. Sci. Arts, 4: 118. 

'^onoUjla glomerata, Cook and Collins, 1895, ]. 
New York Acad. Sci., 9(1): 78. 

The holotNpe, from the John Day \'alley, 
>egon, has been lost, and as Cook and Col- 
ins (1895) stated, the original description 



is too vague to allow definite placement, 
though a certain similarity to Taiyntyla 
emerges. It is probably neither a Tricho- 
petahim nor a Conotyla. I consider it a 
nomen chibium. 

Bollmanello oregona Chamberlin 

BoUmauella oregona Chamberlin, 1941 Bull Univ 
Utah, Biol. Sen, 6(5): 12. 

The male holotype, from John Day 
Creek, Douglas Co., Oregon, is reputedly 
in the Chamberlin collection at the Uni- 
versity of Utah, but it could not be located 
by the curator there. There is nothing 
diagnostic about the description of the 
genus or of the type and only species, 
which even contradict one another in the 
matter of a lobe on the fourth article of the 
fifth legs. No illustrations were presented. 
I consider it a nomen dubium. 

Zygotyla phana Chamberlin 

Zijgotijla phana Chamberlin, 1951, Nat. Hist. 
Misc. No. 87: 7-8, fig. 14. 

The holotype, from blue River, British 
Columbia, Canada, is ob\iously immature, 
having only 28 segments and undeveloped 
gonopods ["The gonopods seem to have 
been broken off in the type (Chamberiin, 
1951)."] Though the type was to have been 
placed in the Provincial Museum of British 
Columbia, they have never received it, and 
it could not be located in the Chamberiin 
collection at the University of Utah. Geo- 
graphically, this species could possibly be a 
synonym of C. atroUneata, but it may also 
represent an undescribed species. Nomen 
dubium. 

Conotyla jonesi Chamberlin 

Conotyla jonesi Chamberlin, 1951, Nat. Hist. 
Misc. No. 87: 7, fig. 1,3. 

The drawing gi\en by Chamberiin 
(1951) for C. jonesi is very much like 
Taiyntyla corvallis when properly oriented; 
the tvpe localities ( t>'pe of T. corvallis from 
Cor\allis, Oregon) are onl\- about 25 miles 
apart and both are in the \Mllamette River 



64 



BiiUetin Museum of Comparative Zoology, Vol 141, No. 2 



valley. None of the eharacters given in the 
original deseription of C. joncsi are in the 
least (liagnostie; they eould apply to 
almost an\' eonotyloid of the region. See 
Hoffman (1961) for a full discussion of 
Chamberlin's errors of observation in the 
deseription of T. corvallis. The type of jonesi 
WAS supposedK' i-)laeed in the Provincial 
Museum of iiritish Columbia, but they 
have no ic^cord of ever having received it, 
and it could not be located in Chamberlin's 
collection at Salt Lake City. At any rate, 
if specimens from the type locality of 
jonesi prove to be identical to corvallis, the 
name may have to be changed to jonesi, 
which has a year's priority. For the present, 
jonesi is best regarded as a nomen duJnum. 

Key to Genera of Conotylidae of 
North America 

la. Anterior ^onopocl sternum divided; anterior 
gonopods fused to each other at base, much 
larger than colpocoxites of posterior gono- 
pods (Fig. 10); northern Illinois, nortlieast- 

ern Iowa, southwestern Wisconsin 

Achemenides 

lb. Anterior gonopod sternum not divided 
(Figs. 1-9); anterior gonopods not fused to 
each other, though they may be closely 
appressed, smaller oi larger than colpo- 
coxites of posterior gonopods 2 

2a. Anterior gonopods flattened, platelike, often 
appressed in the midline (Figs. 8, 9), 
larger than or subequal to colpocoxites of 
posterior gonopods - - 3 

21). Anterior gonopods never flattened, often 
with complex or plumose branches (Figs. 
1, 32-.34), smaller than or subequal to 
colpocoxites of posterior gonopods 4 

3a. Anterior gonopod sternum with lateral 
processes partially fused to lateral sides of 
coxotelopodites (Fig. 69); coxotelopodites 
complex on posterior surface (Fig. 78); 
colpocoxites small, cupped anteriad; Illi- 
nois and Missouri, and Rocky Mountains 
from Alberta to Chihuahua Austrotijla 

3]i. Anterior gonopod sternum without such 
processes, lieavily scleroti/.ed, completely 
surrounding jjases of coxotelopodites ( Figs. 
59, 61, 63); colpocoxites subequal to an- 
terior gonopods, fre(iuently complex and 
branched (Fig. 64), not cupped anteriad; 
Pacific Coast from central Oregon to San 
Francisco Bay region - Taitiiifyhi 

4a. Anterior gonopod sternum with lateral 



process partially fused to lateral edge of 
gonopod; gonopod two-branched, mesal 
edge of major branch heavily laciniate 
(Figs. 80, 81); colpocoxitc two-branched; 
animals without pigment, ocelli irregular; 

caves in northern California Pltimattjla 

4b. AnlcMJor gonopod sternum simple and 
bandlike (Figs. 1-7); gonopod with one or 
two small branches, usually not laciniated, 
colpocoxites with complex posterior surface 
(Fig. 52), but not two-branched; animals 
usually pigmented, ocelli round, black; 
eastern North America from Maine to 
North Carolina, Rocky Mts. of Alberta and 
British Columbia, Sierra Nevada Mts. of 
California Conotyla 

Genus Conotyla Cook and Collins, 1895 

Craspedosoma, Bollman ( in part, not of Leach, 

1815), 1893, U. S. Nat. Mus. Bull. No. 46, pp. 

35, 183. 
Tiichopetalum, McNeill (in part, not of Harger, 

1872), 1887, Proc. U. S. Nat. Mus., Vol. 10, p. 

330; Bull. Brookville Soc, Vol. 3, p. 8. 
Scoterpes, Bollman (in part, not of Cope, 1872), 

1893, Bull. U. S. Nat. Mus. No. 46, p. 106. 
Conotiila Cook and Collins, 1895, Ann. New York 

Acad. Sci., Vol. 9, No. 1, pp. 70-71. Hoffman, 

1961, Trans. Amer. Ent. Soc, Vol. 87, p. 265. 
Proconotyla Verhoeff, 1932, Zool. lahrb. Abt. 

Syst., Vol. 62, p. 501. NEW SYNONYMY. 

Type species of Conotyla, Conotyla 
fischeri Cook and Collins, of Proconotyla, 
P. ])lakei Verhoeff. The generic name is 
a feminine Greek noun, "a cone-shaped 
lump," and refers to the segmental shoul- 
ders. 

Diagnosis. Anterior gonopod sternum 
simple and bandlike, not di\ided. Anterior 
gonopods \'ariously branched or simple and 
acuminate, smaller than or subcciual in size 
to colpocoxites of posterior gonopods. An- 
terior gonopods usually curved posteriad 
of colpocoxites. Colpocoxites of posterior 
gonopods large, complex to simple on pos- 
terior surface. Sternum of posterior gono- 
pods bandlike or conspicuously broadened 
and ovate, spiracles large and prominent, 
tracheal apodemes fused to sterna. Progon- 
opodal legs of males frequently with 
femoral knobs, appearing on all legs in one 
case, and on none in another extreme. 
Species usually pigmented; ocelli 14-23, 



MiLLiPED Family Conotylidae • Shear 



65 





0. 5 mm 







0.5 mm 





Figures 7-14. Anatomy of conotylid male gonopods and of Conotyla blakei. Fig. 7. Conotyla albertana gonopods, an- 
terior view. Fig. 8. Taiyulyla corvallis gonopods, anterior view. Fig. 9. Austrotyla coloradensis gonopods, anterior view. 
Fig. 10. Acfiemen/des pectinatus gonopods, anterior view. Fig. 11. Plumatyla humerosa gonopods, anterior view. Figs. 
12-14. Conofy/o blakei. Fig. 12. Leg 7, posterior view. Fig. 13. Gnathochilarium, ventral view. Fig. 14. Segment 7 of 
male, lateral view (anterior to the left). 



66 Bulletin Museum of Comparative Zoology, Vol 141, No. 2 




Map 1. North America, showing distribution of conotylid genera (some records of immature specimens included); dots, 
Conofy/a spp.; circles, Ausfrotyla spp.; triangles, Plumatyla spp.; solid square, Achemenides pectinatus; open squares, Tai- 
yutyla spp. 



usually round, black, arranged in triangular 
eyepatch in four or five rows. Coxae of 
tenth legs with large coxal glands, not 
lobed; coxae of eleventh legs normal; pre- 
femur of eleventh legs with prominent 
posterior hooks. 

Species. Fifteen known. 

Distribution. See Map 1. Eastern North 
America from Maine through \'ermont, 
New Hampshire and New York to Ohio 
and Indiana in the west and to North Caro- 



lina in the east; center of diversity is ap- 
parently in eastern West Virginia and 
southwestern Virginia; Rocky Mountains of 
the Alberta-British Columbia border; Sierra 
Nevada Mts. of California. 

Key to Species (Males) 

la. Pregonopodal legs only sliglitly more eras- 
sate than postgonopodal legs, without 
knobs; Nevada Co., California extorris 

lb. Pregonopodal legs strongly crassate (Fig. 



MiLLiPED Family Conotylidae • Shear 67 



12) at least leg 4 or 7 with a femoral poorly sclerotized plumose rod (Figs. 4, 

2a. LegVwithafem'oralknob":: 3 ^^' ^f^' ^^^'^'1^^. ^^^''l ^"""^ ^ holhnani 

2b. Leg 4 without a femoral knob, leg 7 with ^^® clearly related, but the resemblance of 

strong femoral knob (Fig. 12); Vermont, ^- extorris to these species is probably a 

New York, Pennsylvania, Maryland, and result of parallelism. 

West \'irginia, often in caves blakei 

3a. Leg 7 without a femoral knob 4 n^^^i I Li i • i\i l rr\ 

3b. Leg 7 with a strong femoral knob ..::;Z: 6 ^°"°^y/° ^'^'^f ' (Verhoeff) 

4a. Legs 2 and 3 with femoral knobs; Virginia Figures 12-18 

^, " celeno FwconoUjla hlakci Verhoeff, 1932, Zool. Jahrb. 

4b. Legs 2 and 3 without femoral knobs 5 Abt. Syst., Vol. 62, p. 501, figs. 33-37. Hoff- 

5a. Leg 5 without a femoral knob; Indiana man and Chamberlin, 1958, U. S. Nat. Mus. 

_ .... hollmani Bull. 212, p. 101. Hoffman, 1961, Trans. Amer. 

5b. Leg 5 with a femoral knob; Virginia ... Ent. Soc, Vol. 87, p. 271. 

venetia Conotyla vaga Loomis, 1939, Bull. Mus. Comp. 

6a. Leg 6 without a femoral knob; British Zool., Vol. 86, pp. 182-183, fig. 10. Chamber- 
Columbia atiolincata lin and Hoffman, 1958, U. S. Nat. Mus. Bull. 

6b. Leg 6 with a femoral knob 7 212, p. 99. NEW SYNONYMY. 

7a. Leg 2 with a femoral knob 8 

7b. Leg 2 without a femoral knob 9 Types. Male holotype of P. hiakei, Mt. 

8a. Leg 1 with a femoral knob; West Virginia Adams, Essex Co., N. Y., wherealjouts 

«K I'Z'T'-^'V'f V'T" r "V;." . ^.'''^" unknown; of C. voga, South Temple 

bb. Leg 1 without a femoral knob; Virginia r^ n ^ r^ r. i • • i 

jfjelimla Cave, Berks Co., Pennsylvania; ni the 

9a. Leg 3 with a strong femoral knob 10 ^^- C. Z., examined. 

9l3. Leg 3 with a weak femoral knob, or lack- Diagnosis. Distinct from other species 

T« !!^^ ^ ^^^^ --: :---: ^ of the ])IaJ<ei group in having an apophvsis 

lOa. Posterior surface of colpocoxite (Fig. 52) on l^rr 7 z' T^in- ^0^ 7,^77.„^«,- 1, i 

„-.i ,^,.^ „i 1 1 \ii ^ 77 / on leg / (Pig. Izj /;o///?2fln? has an apophy- 

with several branches; Alberta ... albeitana ■ ii^ t . i/ 

10b. Posterior surface of colpocoxite (Fig. 29) ^^^ ^"y ^^ leg 4, and extonis, besides 

with a single plumose branch; Kentucky occurring in California, has no pregono- 

elpenor podal leg modifications at all. 

11a. Leg 3 with a weak femoral knob 12 Description of mole from Mt. Equinox, 

lib. Leg 3 without a femoral knob 13 R^v,K,;.i.rf^» r'J \/^ * r u icn 

19., PnU^r^r-nvif^ /vi.r Qwix .-i .... .. Dennington Co., Vermont. Length, IS.O 

i.ia. LvOlpocoxite (l"ig. 38) with an attenuate t-. . i /t-.. ■,,- ,^^ •■ 

apical hook; Virginia aeto "^"''- Eyepatches (Figs. 15, 16) triangular, 

12b. Colpocoxite (Fig. 31) blunt apically; 18 ocelli in 4 rows on both sides. Antennal 

West Virginia smilax articles (Fig. 16) in order of length: 3 5 

^^^' Ohio"'''^ ''"°'' °^ ^^^ ^ ^^""^^^ developed; 4, 2 = 6. 7, \. Leg 7 (Fig. 12) with large, 

13b. FemoraTknobof leg'I'strongZ" ^''''°''''[l capitate, proximal apophysis on femur, 

14a. Colpocoxite (Fig. 36) with an attenuate pregonopodal legs otherwise unmodified. 

distal process; Ohio ocijpetes Anterior gonopods (Fig. LS) prolonged 

14b. Colpocoxite (Fig. 25) apically blunt; directly into posterior branch, anterior 

■ ^ Tit"' • . fi-^chcri branch single, inconspicuous plumose rod; 

The following arrangement into species .^rior branch narro^^^ing at tip, curxing 

gioups IS not wholly a na ural one, but ,,„,,,d lateral side of colpocoxites of posted 

selves the purpose o mabng the task ,^^, gonopods. Colpocoxites of posterior 

e!silr rf ^^^^^^^^'^^^t^d^Pf^^% diagnoses , ^s (Fig. 17) acuminate, drawn out 

easier. The arrangement is based primarily .^.^othlv into decurved terminal process; 

on the form of the anterior gonopods. j^^t belo^^' origin of terminal process is a 

^r ^ T>r .r.r^ o uicsal traHsparcut plate; pilose area limited 

1 HE Blakei Group *. i i. i r -..i ^ i i i 

to lateral surface, without branches; mesal 

This group is characterized by simple, tooth and lateral notch not present. 

usually acuminate anterior gonopods in Remarks. This species is widely dis- 

vvhich the anterior branch is reduced to a tributed (Map 2). It has been collected 



68 



BuUctin Museum of Comparative Zoology, Vol. 141, No. 2 




Figures 15-20. Anatomy of Conofy/o blakei and gonopods of Conotyla spp. Figs. 15-18. Conofy/o blakei. Fig. 15. Head 
and anterior trunk segments of female, lateral view. Fig. 16. Head of female, anterior view. Fig. 17. Colpocoxite of left 
posterior gonopod, posterior view. Fig. 18. Left anterior gonopod of male, posterior view. Figs. 19-20. C. bollmani. Fig. 
19. Left anterior gonopod of male, posterior view. Fig. 20. Left posterior gonopod, posterior view. 



MiLLiPED Family Conotylidae • Shear 69 




Map 2. Part of northeastern United States, showing distribution of some Conofy/o spp.; dots, epigean records of C. blakei; 
circles, cave records of C. blakei; crosses, records of C. fischeri. Question mark indicates a dubious identification. 
Map 3. Part of central United States, showing distribution of Austrotyla and Achemen/des,- dots, epigean records of 
Austrotyla specus; circles, cave records of A. specus; crosses, records of Achemenldes pecttnatus (all cave records). Ques- 
tion mark indicates a dubious identification. 



in the Adirondack counties of New York 
in Canadian Zone Forests and in caves; in 
caves in Sullivan, Orange, and Schoharie 
counties, and in numerous caves in central 
Pennsyh'ania and Maryland, and a single 
case in Jefferson Co., West Virginia. I 
have personally collected blakei in^ a 
scrubby fir forest at 3800 feet on the sum- 
mit of Big Equinox Mountain, Bennington 
Co., Vermont, where it was abundant under 
the bark of both standing and fallen dead 
trees. The distribution of ]}Ia]<ei follows 
exposures of limestone and marble up to 
the Adirondack's, where it is also found on 
other substrates. This distributional pat- 
tern seems to clearly indicate that ])kikei 
followed the boreal forests north with the 
retreat of the Wisconsin ice at the end of 
the Pleistocene, leaving behind relict popu- 
lations in suitable habitats, such as lime- 
stone caves. 



The cave populations described bv 
Loomis (1939) as voga differ hardly at all 
from the typical form of blokei, except that 
the pigmentation is much weaker in some 
specimens and the arrangement and size 
of the ocelli is somewhat irregular. 

Conofyla bollmani (McNeill) 
Figures 19, 20 

TricJiopctahtm hoUmani McNeill, 1887, Proc. U. 
S. Nat. Mus., \'ol. 10, p. 330. 

Scotherpes tvyandotte Bollnian, 1889, Proc. U. S. 
Nat. Mus., Vol. 11, p. 405. Misspelling of 
Scoterpes wijandotte. NEW SYNONYMY. 

Conotijla bollmani. Cook and Collins, 1895, Ann. 
New York Acad. Sci., Vol. 9, p. 76, figs. 
79-94. Loomis, 1943, Bull. Mus. Comp. Zool., 
Vol. 92, p. ,381 (key). Chamherlin and Hoff- 
man, 1958, U. S. Nat. Mus. Bull. 212, p. 98 
(list). 

Conofyla tvyandotte. Cook and Collins, 1895, Ann. 
New York Acad. Sci., Vol. 9, p. 78, fig. 101; 
Chamberlin and Hoffman, 1958, U. S. Nat. 



70 Bulletin Museum of Comparative Zoology, Vol. 141, No. 2 



Mus. Bull. 
SYNONYMY, 



212, p. 100 (list). NEW 



Types. Male types of T. ])oUmani from 
Mayfield's Cave, Bloomington, Monroe 
Co., Indiana; U'pe in U. S. National 
Museum (?); of C. Wyandotte, near 
Wyandotte cave, Crawford Co., Ind.; 
type in U. S. National Museum. 

Dio<!,nosi.'i. Differs from the other species 
of the hlokei group in having an apophysis 
on the fourth article of the fourth leg; 
hhkei and extorris both lack such an 
apophysis. 

Description of male from Boone Cove, 
Owen Co., Indiana. Length, 14.0 mm. 
Ocelli 22 in four rows plus single ocellus 
on the right side, 20 ocelli in four rows on 
left side. Antennal articles in order of 
length, 3, 4, 5, 2, 6, 7, 1. Body white, 
ocelli somewhat irregular in shape and ar- 
rangement. Leg 4 with large capitate 
apophysis on femur. Anterior gonopods 
( Fig. 19 ) simple, acuminate to tip of poste- 
rior branch, anterior branch a simple, 
inconspicuous plumose rod; lateral shoulder 
low, rugose. Anterior gonopods passing 
around lateral side of colpocoxites of poste- 
rior gonopods in situ. Colpocoxites of 
posterior gonopods (Fig. 20) bluntly 
pointed, without definite terminal process; 
mesal tooth bifid; pilose area small, re- 
stricted to part of posterior surface. 

Notes. Known primarily from caves and 
a few surface records in south-central In- 
diana. Surface specimens are normally 
pigmented, while cave populations are 
usually white and have rather irregular 
ocelli. The identity of immature and fe- 
male specimens from adjacent regions of 
Kentucky is unclear. Map 7 shows the 
boundaries of the supposed glacial luaxima 



of Wisconsin and of Illinois time (from 
Wayne and Zumberge, 1965), but since 
the availability of a suitable substrate also 
affects the distribution of cave animals, and 
surface records are scarce, I would rather 
not speculate on the possible histoiy of this 
species. 

Conotyla extorris n. sp. 
Figures 22, 23 

Type. Male holotype from 1 mile south 
of Grass Valley, Nevada Co., Cali- 
fornia, collected January 5, 196S, by 
D. E. Bragg and R. F. Denno. The 
specific epithet is a Latin adjective 
meaning "banished, exiled," and refers 
to the distance separating this species 
from the others of its genus. 
Dia<i,nosis. Distinct from others of the 
hlakei group in having no modifications at 
all on the pregonopodal legs, which are of 
normal size and lacking apophyses. The 
gonopods (Figs. 22, 23) are only about a 
third the size of those of hoUmani and 
hlakei. 

Description of holotype male. Length, 
12.6 mm, the smallest known species of 
Conotyla. Eyepatches triangular, 20 ocelli 
in four rows on the right side, 19 ocelli in 
four rows on the left side. Antennal joints 
in order of length: 3, 5, 4, 6, 2, 7, 1. Pre- 
gonopodal legs unmodified. Anterior gono- 
pods ( Fig. 22 ) with posterior branch large, 
broadened near the tip and recurved, pass- 
ing laterad of the colpocoxites of the poste- 
rior gonopods; anterior branch a single, 
inconspicuous plumose rod. Colpocoxites 
of the posterior gonopods (Fig. 23) with 
terminal process sharply curved dorsad and 
slightly laterad; mesal tooth (?) a long rod 
reaching to the tip of the colpocoxitc; 
lateral notch and pilose area absent. 



Figures 21-29. Male gonopods of Conofy/o spp. Fig. 21. Conotyla bollmani, left anterior gonopod, lateral view. Figs. 
22-23. C. exforris. Fig. 22. Right gonopods, anterior view. Fig. 23. Right gonopods, posterior view. Figs. 24-25. C. 
fischeri. Fig. 24. Left anterior gonopod, posterior view. Fig. 25. Left posterior gonopod colpocoxite, posterior view. Figs. 
26-27. C. personafo. Fig. 26. Left anterior gonopod, posterior view. Figs. 28-29. C. elpenor. Fig. 28. Left anterior gono- 
pod, posterior view. Fig. 29. Colpocoxite of left posterior gonopod, posterior view. 



MiLLiPED Family Conotylidae • Shear 71 







72 Bulletin Mtisniw of Comparaiivc Zoologij, Vol. 141, No. 2 



Notes. The type locality (Map 6) is 
near the Boyce Thompson Institute of 
Forest Studies, at about 2400 ft. ele\ation. 
The area is forested primarily with pine, 
Arctostciphylos, and oaks. Michael Gardner 
(personal communication) has recorded the 
millipeds Bmchiicyhe producta, Wamolda 
sierra. Buzoniiim crassipes, and Phicerna 
dorada, among others, from Grass Valley. 

The assignment of this species to 
Conotyla may prov(^ to be controversial. 
It has many of the characters of Taiytitylo. 
a genus gt>ographicall\' closer. HoNxt^ver, 
the form of the gonopods and especially the 
anterior gonopod sternum favor the present 
placement. If more new species with a 
similar combination of characters are col- 
lected, a new genus may be justified. 

The Fischeri Group 

This is a compact, closely related group 
of three species from New York, Ohio, and 
Kentucky, characterized by the reduction 
of the anterior branch of the anterior 
gonopods, and by their similar colpocoxites. 

Conotyla fischeri Cook and Collins 
Figures 3, 24, 25 

Conotifla fischeri Cook and Collins, 1895, Ann. 
New York Acad. Sci., Vol. 9, pp. 71-74, figs. 
55-78. Loomis, 1943, Bull. Mus. Comp. Zool., 
Vol. 92, p. .382 (key). Hoffman, 1961, Trans. 
Amer. Ent. Soc, Vol. 82, pp. 265-266, fig. 7. 
Chamberlin and Hoffman, 1958, U. S. Nat. 
Mns. Bull. 212, pp. 98-99 (list). 

Types. None designated by original 

authors; type locality restricted by 

Chamberlin and Hoffman ( 1958 ) to 

Syracuse, Onondaga Co., New York. 

Diagnosis. Distinct in the form of the 

gonopods from all other species except the 

closely related personata; differing from 

personota in the gonopods as described 

under that species, and also in having a 

much more pronounced apo^jhysis on the 

femur of leg 4. 

Description of mole from Skaneateles, 
Onondaga Co., New York. Length, 19.4 



inm. Eyepatches triangular, 20 ocelli in 
four ro\x's on left side, 21 ocelli in four 
rows on right side. Antennal segments in 
order of length: 3, 4 = 5, 2, 6, 7, 1. Pre- 
gonopodal legs somewhat less enlarged 
than usual, legs 4-7 usually with apophyses 
on femur; distal on leg 4, mesal on legs 5 
and 6; large, proximal and toothed, if pres- 
ent, on leg 7, but not resting on anterior 
gonopod in situ. Anterior gonopods (Fig. 
24) short, subfiuadrate, posterior branch 
absent, anterior branch short, not curving 
around colpocoxites of posterior gonopods; 
lateral shoulder poorly developed. Colpo- 
coxites of posterior gonopods (Fig. 25) 
without a terminal process, but with a 
prominent ridge on the posterior surface, 
bearing near its proximal end a long, rather 
flattened rod; lateral notch absent; pilose 
area with one or two pilose branches. 

Notes. Cook and Collins (1895) had 
over 100 specimens from Onondaga, On- 
tario, and Wayne counties, New York 
(Map 2). They were collected primarily 
from rotting littcn- in woods, and under the 
moist, rotting bark of fallen trees. Hoffman 
( 1961 ) reported the species from Hamilton 
Co., 2 miles west of Morehouseville, but 
gave no additional ecological data. His 
material, which I re-examined for this 
study, differs from the Cook and Collins 
series and from other Onondaga Co. speci- 
mens in having the apophysis of the seventh 
legs very small, or absent; the other 
apophyses are normally developed. The 
species is evidently limited to the rolling 
hills south and east of Lake Ontario (Map 
2), but extension of its range into areas 
bordering on Lake Erie would be most 
interesting, as it might indicate the exact 
nature of this species' relationship to per- 
sonata, which occurs in similar terrain just 
south of Lake Erie in Ohio. Several in- 
triguing immature specimens from Ontario, ' 
northern New York, and from Potter Co., 
Pennsylvania, this latter locality near the 
southern limit of Pleistocene glaciation, 
may belong to this species. 



MiLLiPED Family Conotylidae • Shear 



73 



Conotyla personata n. sp. 
Figures 26, 27 

Type. Holotype male collected by A. A. 
Weaver, May 5, 1960, Funk's Hollow, 
Wooster, Wayne Co., Ohio. The spe- 
cific epithet is a Latin adjective, 
meaning "masked, hidden," and refers 
to my original confusion between this 
species and fisclieri. 
Diagnosis. Very similar to fisclieri, but 
distinct in details of the gonopods. The 
lateral shoulder of the anterior gonopod is 
much higher (Fig. 26) than in fisclieri, 
and the posterior branch is longer and 
more pronounced; the lateral notch of the 
colpocoxite is almost obsolete in fisclieri; 
in personata it is the deepest and broadest 
of any species in the genus. 

Description of male holotype. Length, 
14 mm. Ocelli in four irregular rows on 
each side, 21 on right side and 23 on left 
side. Antennal segments in order of length: 
3, 4, 5, 6 = 2. 7, 1. Fourth joints of legs 4, 
5, and 6 with small, distal knob; fourth 
joint of leg 7 \\'ith long, apically toothed 
proximal process that in situ rests on the 
lateral shoulder of anterior gonopod. Ster- 
num of anterior gonopods as described for 
genus, but lateral edge not visible in poste- 
rior view. Anterior gonopods (Fig. 26) 
short, subrectangular in posterior view; 
lateral shoulder prominent; anterior branch 
absent, posterior branch curved dorsad 
slightly. Colpocoxite of posterior gonopod 
(Fig. 27) bluntly subtriangular, all proc- 
esses and teeth undeveloped; pilose area 
extended into at least one pilose branch; 
lateral notch deep, extending nearly half- 
way up colpocoxite. 

Notes. The male holotype was taken in 
a deep, heavily \^'ooded, mesic ravine sur- 
rounded by rolling cultivated land. The 
north-facing slope of the ravine supports a 
heavy mixed forest with much hemlock; 
the south-facing slope is forested with oak 
and pine. A study of the humus fauna of 
these slopes produced several immature 
and female Conotyla in early fall. This 



species is possibly widespread in northern 
and central Ohio, l^ut the only males are 
from the type locality, and nearby. 

Conotyla elpenor n. sp. 
Figures 28, 29 

Type. Male holotype collected with an- 
other male and two females on 26 
February 1966 by Branley A. Branson, 
nine miles northwest of Pine Ridge, 
ekn'ation 800 ft. (250 m), Wolfe 
County, Kentucky. The specific epithet 
refers to one of the companions of 
Odysseus, changed into a swine by 
Circe's witchcraft. 
Diapjnosis. With the characters of the 
fisclieri group; distinct in the form of the 
anterior gonopods, with two well-defined 
branches ( Fig. 28 ) , and in the colpocoxites 
of the posterior gonopods having an ex- 
ceptionally deep lateral notch and coxal 
pocket. 

Description of holotype male. Length, 
18 mm. Eyepatches subtriangular, 20 ocelli 
in four rows on each side. Antennal articles 
in order of length: 3, 4 = 5, 6, 2, 7, 1. Pre- 
gonopodal legs 3 through 7 with apophyses 
on femur; apophysis of leg 3 small, distal; 
of legs 4 and 5 mesodistal, larger; of leg 
6, proximal and blunt, of leg 7 proximal 
and greatly elongated, coarsely toothed 
distally, resting in situ posterior of lateral 
shoulder of anterior gonopod. Anterior 
gonopod (Fig. 28) with lateral shoulder, 
large, lamellate; two branches of gonopod 
may represent a bifurcation of the posterior 
branch, in situ they extend laterally 
around colpocoxites of posterior gonopods. 
Colpocoxites of posterior gonopods (Fig. 
29) closely resembling those of personata; 
but somewhat longer, bent sharply pos- 
teriad, blunt; lateral notch deep, pilose 
area with two indistinct branches. 

Notes. The holotype and the other 
known specimens were taken from beneath 
a rotting log in mixed hardwood forest. 
Though the specimens were labelled as 
being from Wolfe Co., the distance north- 



74 Bulletin Museum of Comparative Zoology, Vol. 141, No. 2 




MiLLiPED Family Conotylidae • Shear 



75 



west of Pine Ridge indicated on the label 
would place the locality in Powell Co. This 
species is known only from the type lo- 
cality. A female from Carter Co., Kentucky 
(MCZ), may belong to elpenor (Map 5). 

The Smilax Group 

The single species of this group, C. 
smilax, is unique in several ways, but may 
be related to the preceding group. 

Conotyla smilax n. sp. 
Figures 30, 31 

Ttfpe. Holotvpe male collected ca. elev. 

'3000 ft. (ioOO m). Kate's Mt., above 

White Sulphur Springs, Greenbriar 

Co., West Virginia, April 1, 1967, by 

W. A. Shear. The specific epithet is a 

Latin noun in apposition and refers to 

Smilax, the greenbriar, abundant at 

the type locality. 

Diap,nosis. Distinct from all other related 

species of Conotyla in lacking the posterior 

branch of the anterior gonopods ( Fig. 30 ) . 

The anterior branch is well developed and 

had a deeply laciniated membrane running 

the length of its posterolateral edge. 

Description of male holotype. Length, 
13.3 mm. Ocelli in four ro\\'s on each side, 
20 ocelli in each eyepatch. Legs 3 through 
7 enlarged and crassate, knob present on 
femur of legs 3 through 7. Antennal articles 
in order of length: 3, 4 = 5, 6 = 2, 7, 1. 
Sternum of anterior gonopods as described 
for genus. Anterior gonopods (Fig. 30) 
with conspicuous lateral shoulder, but knob 
of fourth segment of leg 7 does not rest 
on shoulder when gonopod is in situ; poste- 
rior branch absent, anterior branch large, 
thin, nearly half as long as remainder of 
coxotelopodite, extending straight ventrad, 
with a deeply laciniated membrane on 



posterolateral surface. Colpocoxites of 
posterior gonopods with terminal process 
long, aciculate; mesal tooth inconspicuous, 
pilose area with definite branches, ex- 
tensive (Fig. 31). 

Notes. The male holotype, the only 
known specimen, was collected after a 
light rain, crawling about on dead twigs 
under a Rhododendron. The type locality 
(Map 5) is an unusual area, dissected into 
adjacent areas of very mesic forest of mixed 
hardwoods and hemlock and "shale bar- 
rens," with a hot, diy microclimate and 
sparse vegetation of scrub oak and Vir- 
ginia pine. Apheloria trimaculata and 
Cleidogonu major were taken in abundance. 

The Venetia Group 

These species have similar anterior gono- 
pods in which the anterior branch is absent 
and the posterior branch long, giving a 
right-angled appearance to the gonopod 

{kg. 33). 

Conotyla venetia Hoffman 
Figures 32-34 

Conottila veitetia Hoffman, 1961, Trans. Amer. 
Ent. Soc, Vol. 87, p. 267, Pi. IX, figs. 1-3. 

Type. Male holotype from three miles 

north of Clifton Forge, Alleghany Co., 

Virginia; type in U. S. National 

Museum, examined. 

Diagnosis. Distinct from all other species 

of Conotyla in bearing apophyses on the 

fourth article of legs 4 and 5 only; distinct 

in details of the gonopods from others of 

the venetia group. 

Description of male from Warm Spring 
Mt., Alleghany Co., Virginia. Length, 14.5 
mm. Ocelli 20 on each side, in four rows. 
Antennal segments in order of length: 3, 
4, 5, 2, 6, 7, 1. Legs 4 and 5 with small 



<- 

Figures 30-37. Male gonopods of Conotyla spp. Figs. 30-31. Conotyla smi/ox. Fig. 30. Left anterior gonopod, posterior 
view. Fig. 31. Left posterior gonopod, posterior view. Figs. 32-34. C. venetia. Fig. 32. Left anterior gonopod, lateral 
view. Fig. 33. Colpocoxite of left posterior gonopod, posterior view. Fig. 34. Coxae of 10th legs of mole, ventral view. Figs. 
35-36. C. ocypefes. Fig. 35. Left anterior gonopod, posterior view. Fig. 36. Left posterior gonopod, posterior view. Fig. 
37. C. oeto, left anterior gonopod, posterior view. 



76 Bullet in Museum uf Comparative Zoology, Vol. 141, No. 2 



apophxses on femora, distal in position. 
Anterior gonopods (Figs. 32, 33) with 
posterior branch extending at right angle 
posteriad from t(>lopodite, anterior branch 
absent; lateral shoulder well dexeloped, 
heavily rugose on latc^ral surfaces posterior 
branch extending in situ well around lateral 
surface of colpocoxite of posterior gonopod. 
Colpocoxite of posterior gonopod (Fig. 34) 
subrectangular in outline; terminal process 
definitely set off by promintnit shoulders 
at its base; mesal tooth reduced to small 
knob; pilose area lateral; lateral notch 
present, but small. 

Notes. This species is known only from 
two localities in Alleghany Co., Virginia 
(Map 5), separated bv nearlv 3000 ft. 
(1000 m) of altitude ( 400-3400' ft. ) . Hoff- 
man (1961) implies that the species may 
be widespread in central western Virginia. 

Conofyla ocypefes n. sp. 
Figures 35, 36 

Type. Male holotype from Sugar Grove, 
Fairfield Co., Ohio, collected by J. H. 
Emerton, December 26, 1915. The spe- 
cific epithet is a noun in apposition, 
the Greek name of one of the three 
Harpies of mythology. 
Dia<inosis. Distinct from others of the 
venetia group in details of the gonopods 
and in the apophyses of the pregonopodal 
legs; from aeto in having a strong apophysis 
on the third joint of leg 2 and none on leg 
3, and from venetia by having apophyses 
on legs 6 and 7. 

Description of liolotype male. Length, 
20.0 mm. Eyepatch of right side triangular- 
truncate, 20 ocelli in four rows; eyepatch 
of left side probably anomalous, five large 
ocelli in two rows. Antennae missing. Leg 
2 with strong distal apophysis on third 
joint; legs 4 through 6 with apophyses on 
femora becoming smaller and mesal; leg 
7 with small proximal apophysis on femur 
that does not touch anterior gonopod in 
situ. Anterior gonopod (Fig. 35) typical 
of venetia group, anterior branch absent. 



posterior branch large, becoming spatulate 
distally, curving around lateral side of 
colpocoxites of posterior gonopods; lateral 
shoulder not at all developed, instead 
gonopod is highly ridged mesally. Colpo- 
coxites of posterior gonopods ( Fig. 36 ) un- 
usually long, but arched dorsad; terminal 
process very long and irregularly sinuate, 
with a small triangular tooth about half its 
length from the origin; mesal tooth present; 
lateral notch absent; pilose area small, 
limited to lateral margin. 

Notes. The male holotype was collected 
with a female and an immature specimen; 
probably all are of the same species. The 
southern section of Fairfield County ( Map 
5) is in the valley of the Hocking River, a 
region of low hills highly dissected into 
deep gorges with abrupt sandstone walls. 
Most of these gorges are very moist; the 
vegetation is mostly hemlock and beech. 
Emerton unfortunately failed to indicate 
exactly where the type specimen was ob- 
tained, but it seems likely that such a gorge 
as the ones described above would provide 
the most favorable habitat. However, 
several years of extensive collecting in the 
Hocking area by Dr. A. A. Weaver and 
others has resulted in the collection of no 
conotylids. 

Conofyla aeto n. sp. 
Figures 37, 38 

Type. Male holotype from top of Clinch 
'Mt, ca. 4500 ft. (1500 m), Burke's 
Garden, Tazewell Co., Mrginia, col- 
lected November 14, 1965, by Radford 
College Biology Club. The specific 
epithet, like celeno and ocypetes, is a 
Greek noun in apposition, the name of 
one of the Hai-pies of mythology. 
Diagnosis. Distinct from other species of 
the venetia group in the modifications of 
the pregonopodal legs and in details of the 
gonopods; from venetia in having apophy- 
ses on legs 6 and 7, and from ocypetes in 
having a slight apophysis on leg 3 and none 
on leg 2. 



MiLLiPED Family CoNOTi'LiDAE • Shear 77 




Map 4. Colorado, dots showing distribution of Austrofyla coloradensis. 

Map 5. North central Appalachian region and part of Ohio, showing distribution of Conotyla spp.; dots, C. mel/ndo; 
circles, C. venetio; cross, C. ocypefes; a, C. aeto; c, C. ce/eno; e, C. e/penor; v, C. vista; s, C. smi/ax. Question mark in- 
dicates a dubious identification. 



Description of liolotype mole. Length, 
15 mm. Eyepatch triangular, 21 ocelli in 
four rows and single ocellus on right side; 
20 ocelli with same arrangement on left 
side. Antennal articles in order of length: 
3, 4, 5, 2 = 6, 7, 1. Legs 3 through 7 with 
apophyses on femora, that of leg 3 small, 
becoming stronger to leg 6, all distal in 
position; apophysis of leg 7 proximal in 
position, capitate and toothed, but not 
resting on shoulder of anterior gonopod 
in situ. Prefemoral hooks of leg 11 particu- 
larly large. Anterior gonopods (Fig. 37) 
typical of venetia group, anterior branch 
absent, posterior branch large, broadened 
towards tip, extending well around lateral 
side of colpocoxites of posterior gonopods 
in undissected animal; lateral shoulder 
poorly developed. Colpocoxites of posterior 
gonopods (Fig. 38) distinct; terminal proc- 
ess flattened dorsoventrally, hooked dor- 
sad; subterminal process expanded laterad 
into a large plate that in natural ^Dosition 
covers apical end of posterior branch of 
anterior gonopod and is heavily ridged on 
posterior side, the distal ends of the ridges 
drawTi out into prominent teeth (Fig. 38); 



mesal tooth absent; pilose area with single 
pilose branch; lateral notch absent. 

Notes. The northeastern part of Clinch 
Mountain, which extends from southwest 
to northeast across western Virginia, forms 
the southern and western wall of Burke's 
Garden, an unusual semicircular anticlinal 
valley, the floor of which is at nearly 3000 
ft. ( 1000 m ) elevation, surrounded l^y 
mountains up to 4800 ft. (1600 m), with 
their tops in well-developed Canadian Zone 
forests. 

The Melinda Group 

Both branches of the anterior gonopod 
are nearly equal in this group. The poste- 
rior branch is distally plumose, but strong 
and heavy. 

Conofyla melinda Hoffman 
Figures 1, 2, 39, 40 

Conotyla melinda Hoffman, 1961, Trans. Amer. 
Ent. Soc, Vol. 87, p. 266, PL IX, figs. 4-6. 

Type. Male holotype from Brush Mt., 
2 miles west of Blacksburg, Montgom- 
ery Co., Virginia; type in U. S. 
National Museum, examined. 



78 BiiUctin Muscunt of Comparative Zoology, Vol. 141, No. 2 




MiLLiPED Family Conotylidae • Shear 



79 



Diagnosis. Distinct from others of the 
melinda group in details of the gonopods 
and the distribution of apophyses on the 
pregonopodal legs; vista has apophyses on 
all the pregonopodal legs, celeno lacks 
them on leg 6. 

Description of male from type locality. 
Length, 18.5 mm. Ocelli 20 in triangular 
eyepatch of four rows on right side, 19 
ocelli in four rows on left side. Antennal 
articles in order of length: 3, 5, 4, 2, 6, 7, 1. 
Leg 2 with large apophysis on third joint; 
legs 3 through 6 with similar apophyses on 
femora, largest on legs 4 and 5, distal in 
position. Anterior gonopods ( Fig. 39 ) with 
anterior branch larger than posterior, 
twisted and deeply laciniate at the end; 
posterior branch a short hook; lateral 
shoulder with prominent knobs. Colpo- 
coxites (Fig. 40) of posterior gonopods 
with terminal j)rocess bluntly triangular, 
curved posteriad, pilose area extensive; 
mesal tooth absent, lateral notch inconspic- 
uous. 

Notes. Hoffman's original description 
overlooked the posterior branch of the 
anterior gonopod. Reported from Craig, 
Montgomery, Giles and Patrick counties 
in southwest Virginia ( Map 5 ) . All records 
are for November and December, except 
for a pair taken in copula on March 15, 
1956, in oak woods. Other records are from 
Rhododendron thickets and a sinkhole 
(Hoffman, 1961). 

Conotyla vista n. sp. 
Figures 41, 42 

Type. Male holotype collected by W. A. 
Shear and David Bard, Januarv 25, 
1967, Natural Tunnel No. 1, Grand- 
view State Park, 13 miles northeast of 
Beckley, Raleigh Co., West Virginia, 
elevation 2250 ft. (710 m). The spe- 



cific epithet is a Latin noun in appo- 
sition and refers to tlie sweeping view 
of the New River canyon visible from 
the type locality. 
Diagnosis. Distinct from the other species 
of the melinda group, and from all other 
known Conotyla species, in having promi- 
nent apophyses on all the pregonopodal 
legs, including the first. 

Description of holotype male. Length, 
26 mm, the largest known species of the 
family. Nineteen ocelli in four rows in a 
triangular eyepatch on left side, 16 ocelli 
in four rows on right side. Antennal joints 
in order of length: 3, 4, 5, 2, 6, 7, 1. Joint 
3 of legs 1 and 2 with mesal apophyses of 
moderate size; similar apophyses on the 
femora of legs 3 through 6, becoming 
largest on leg 4; femur of leg 7 with an 
exceptionally large, apically toothed, sinu- 
ate process on proximal end, that in situ 
rests in pocket, on anterior gonopod of 
each side (Fig. 41). Anterior gonopods 
with lateral shoulder prominent and 
rimmed; mesal to shoulder apical region of 
gonopod is depressed, forming a pocket 
in which apophysis of leg 7 fits when 
gonopods are in situ; posterior branch 
somewhat larger than anterior, somewhat 
laciniate apically, both branches pass later- 
ally around the colpocoxites in situ. Colpo- 
coxites of posterior gonopods (Fig. 42) 
with a long, sigmoid and aciculate apical 
process, pilose area occupying the lateral 
margin; lateral notch and mesal tooth 
absent. Colpocoxite distinctly thickened on 
mesal margin. 

Notes. The male holotype is the only 
known specimen. The type locality (Map 
5) is a ridge of heavilv faulted, coarse 
sandstone overlooking the 1200 foot deep 
gorge of the New River. The Natural Tun- 
nels are roofed crevices formed by down- 



Figures 38-43. Male gonopods of Conotyla spp. Fig. 38. Conotyla aeto, colpocoxite of left posterior gonopod, posterior 
view. Figs. 39-40. C. melinda. Fig. 39. Left anterior gonopod, posterior view. Fig. 40. Colpocoxite of left posterior 
gonopod, posterior view. Figs. 41-42. C. vista. Fig. 41. Left anterior gonopod and basal segmenis of leg 7, posterior 
view. Fig. 42. Left posterior gonopod, poslerior view. Fig. 43. C. celeno, left anterior gonopod, posterior view. 



80 



BuUetin Muscimi of Comparalivc Zoology, Vol. 141, No. 2 



slopr crcepins; of sandstone blocks, and are 
long enough to ha\e totalK* dark areas and 
at least some troglophilie species (Meta 
menardii and Caiymmaria cavicola, both 
spiders), but C. vista shows no cave modi- 
fications. Folding-door spiders (Antrodioe- 
tiis sp. ) are very common in the immediate 
area. 

Conofyla ce/eno n. sp. 
Figures 43-45 

Type. Holotype male collected Decem- 
ber 12, 1965, bv R. L. Hoffman, in 
vicinity of Comer's Rock, Iron Mt., ca. 
4000 ft. (1300 m), Grayson-Wythe 
COS., Virginia. The specific epithet is 
a noun in apposition, and is the Greek 
name of one of the three Harpies of 
ancient mythology. 
Diagnosis. Distinct from the other species 
of the melinda group in details of the 
gonopods, but is most easily separated 
from vista by the absence in celeno of an 
apophysis on the third joint of the first leg, 
and from melinda by the absence of 
apophyses on legs 6 and 7. 

Description of holotype male. Length, 
23.8 mm. Ocelli in triangular patch, in four 
rows plus single ocellus on right side, total 
of 19; in five rows on left side, total of 21. 
Antennal joints in order of length: 3, 4, 5, 
2, 6, 7, 1. Leg 2 with distinct distal 
apophysis on the third joint; apophyses 
mesal in position on the femora of 3 and 4, 
apophyses of leg 4 the largest, distinctly 
capitate; apophysis of femur of leg 5 very 
small; legs 6 and 7 unmodified. Anterior 
gonopods (Figs. 43, 44) with lateral 
shoulder of normal size, lateral surface of 
lateral shoulder heavily pebbled; anterior 
and posterior branches subequal in size, 
anterior branch in situ curved around 
colpocoxite of posterior gonopod, anterior 
branch extending straight ventrad, heavily 
laciniated. Colpocoxites of posterior gono- 
pods (Fig. 45) with terminal process thick, 
curv^ed dorsal; lateral margin flared out as 
large, thin plate, lateral notch not at all 



developed; pknnose area with a single 
branch; mesal tooth large, curved. 

The Atrolineata Group 

This group of two species may be generi- 
cally distinct from Conotyla, but in my 
opinion are not sufficiently differentiated, 
although they have ob\iously had a long 
separate histoiy — perhaps since the Plio- 
cene droughts brought the Great Plains 
into being. They are characterized by 
reduced anterior gonopods, complex colpo- 
coxites, and broadened posterior gonopod 
sterna. 

Conofyla atrolineata (Bellman) 
Figures 6, 46-50 

Craspedosoma atrolineatum Bollman, 1893, Bull. 

U. S. Nat. Mus., Vol. 46, pp. .35-36. 
Conotyla atrolineata. Cook and Collins, 1895, Ann. 

New York Acad. Sci., Vol. 9, p. 75, figs. 95-100. 

Chambeilin and Hoffman, 1958, U. S. Nat. 

Mus. Bull. 212, p. 98 (list). 

Type. Male specimen from Glacier, 
British Columbia; assumed to be Gla- 
cier National Park, although the park 
was not established at the time the 
paper by Bollman was written ( 1887, 
1893 dates the posthumous commemo- 
rative collection of Bollman's work). 
There is a Glacier post office in the 
park on the Canadian Pacific Railroad 
at the head of the Illecillewaet River. 
Type in U. S. National Museum, ex- 
amined. 
Diagnosis. Differing in the expanded 
sternum of the posterior gonopods from all 
species except albertana; from alhcrtana in 
the much larger anterior gonopods and 
details of the colpocoxites of the posterior 
gonopods. 

Description of male from Yoho National 
Park, B. C. Length, 15.5 mm. Eyepatches 
triangular, 20 ocelli in four rows on each 
side. Antennal segments in order of length : 
3, 4 = 5, 2, 6, 7, 1. Leg 4 with large 
apophysis on femur, pregonopodal legs 
otherwise unmodified. Anterior gonopods 



MiLLiPED Family Coxotvlidae • Shear 81 




Figures 44-50. Male gonopods of Cono/y/o spp. Fig. 44-45. Conofy/o ce/eno. Fig. 44. Left anterior gonopod, lateral view. 
Fig. 45. Left posterior gonopod, posterior view. Figs. 46-50. C. atrolineata. Fig. 46. Rigfit anterior gonopod, anterior view. 
Fig. 47. Left posterior gonopod, posterior view. Figs. 48-50. Apical region of colpocoxite of posterior gonopod. Fig. 48. 
Posterior view. Fig. 49. Lateral view. Fig. 50. Mesal view. 



82 



Bulletin Museum of Comporaiwc Zoology, Vol. 141, No. 2 



(Fig. 46) with lateral shoulder laeking; 
large mesal lamella (posterior braneh?) 
present; anterior braneh eui-ving around 
lateral surfaee of posterior gonopods in situ, 
tip three-pointed. Colpocoxites (Figs. 47- 
50) deeply cleft apieally, posterior surfaee 
\\'ith two decur\'ed lamellae and two 
prominent pomted branches; lateral pilose 
area present; mesal tooth sharp, pointed. 

Notes. Known from several localities in 
the Rocky Mountains of eastern British 
Columbia; elevations between 4000 ft. and 
5000 ft. in Yoho National Park, and from 
Robson Creek, Mt. Robson National Park 
(elevation unknown), in addition to the 
type locality. The continental divide sepa- 
rates this species from C alhertana, and it 
seems likely that they took refuge from 
the Cordilleran ice sheets on their present 
sides of the mountains, although there is 
no evidence of glacial refugia, other than 
possible isolated nunataks, in British Co- 
lumbia. 

Conotyla albertana Chamberlin 
Figures 51-53 

Conotyla alhertana Chamberlin, 1920, Canadian 
Ent., Vol. 52, p. 167, fig. 17. Chamberlin and 
Hoffman, 1958, U. S. Nat. Mns. Bull. 212, p. 
98 (hst). 

Type. Male holotype from Bow River, 
Alberta, Canada; in Mus. Comp. Zool., 
examined. The Bow River flows from 
Bow Lake, just south of Bow Pass 
(6878 ft.) through Calgary, Alta., and 
joins the Oldman River to form the 
South Saskatchewan just north of Bow 
Island, Alta., a total course of more 
than 300 miles. Since most other 
known records for C. albertana are 
some 250 miles to the north of the 
river, it seems wise to here restrict the 
type locality to the banks of the Bow 



River in the vicinitv of Lake Louise, 
Banff National Park, Alberta. 

Diagnosis. Distinct from all but atro- 
lineata in having the posterior gonopod 
sternum large, expanded and suboval, with 
prominent spiracles on the anterior surface, 
and in having the colpocoxites of the poste- 
rior gonopods with complex posterior sur- 
faces. From atrolineata, albertana is distinct 
in the two-branched anterior gonopods, as 
described below. 

Description of male from Jasper National 
Park, Alberta. Length, ILO mm. Eyepatches 
subhexagonal (truncate-triangular), five 
rows containing 18 ocelli on right side, 20 
ocelli in five rows on left side. Antennal 
articles in order of length: .3, 4 = 5, 2, 6, 
7, 1. Pregonopodal legs with slight apoph- 
yses on the femora of legs 3, 4, 5, 6; all 
mesal in position, heaviest on leg 4; 
apophysis of leg 6 very small. Anterior 
gonopods (Fig. 53) with lateral shoulder 
so well developed as to appear sigmoid; 
lateral shoulder rugose; anterior branch a 
heavy lamella, pressed against lateral side 
of sternum of posterior gonopod in situ; 
posterior branch cui'ved, rodlike, sharply 
pointed. Posterior gonopods (Figs. 51, 52) 
with colpocoxites truncate, slightly curved 
posteriad, posterior surface with a complex 
group of smooth and plumose branches. 

Notes. Aside from the type locality, this 
species is known from numerous specimens 
from several localities on Mt. Edith Cavell, 
including an alpine meadow, in Jasper 
National Park, Alberta, and Sulphur Mt. in 
Banff National Park. Unfortunately, the 
exact locality of some of the Jasper speci- 
mens was not indicated on the labels, only 
the altitudes (4500, 5000, and 5300 ft.). 
Since the tree line in wc\stern Alberta at 
the latitude of the park is from 7000-7500 



Figures 51-55. Male gonopods of Conotyla and Achemenides. Figs. 51-53. Conotyla albertana. Fig. 51. Right gonopods, 
anterior view. Fig. 52. Left gonopods, posterior view. Fig. 53. Right anterior gonopod, anteromesal view. Figs. 54-55. 
Achemenides pectinatus. Fig. 54. Right anterior gonopod and sternum, lateral view. Fig. 55. Basal region of anterior gono- 
pods, anterior view. 



MiLLiPED Family Conotylidae • Shear 83 




84 Bidletm Museum of Comparative Zoology, Vol 141, No. 2 



ft., depending on loeal eonditions, it seems 
safe to assume that alhertana inhabits the 
Canadian and Hudsonian Zones. During 
the Wisconsin glaciation, there may have 
been an ice-free refugial corridor between 
the Cordilleran and Keewatin ice sheets 
(Moss, 1955), into which albeiiana or its 
ancestral form may have retreated, but Httle 
is known of the vegetation of this corridor. 
From such a refugium, alhertona may have 
re-invaded the montane regions. 

Chamberhn's male holotype lacks the 
seventh legs; he failed to see the apophysis 
of leg 6. The posterior branch of the an- 
terior gonopods is also missing in the holo- 
type. 

Genus Achemenides new genus 

Type species. Conotyh pectinata Causey. 
The generic name (Greek, masculine) re- 
fers to a Greek marooned in a cave after 
the Trojan War. 

Diagnosis. Medium-sized conotylids with 
somewhat reduced ocelli (14-17 in the 
single known species). Antennal segment 
5 nearly twice as long as segment 4. Pre- 
gonopodal legs of male with apophyses on 
either third or fourth segments or both. 
Anterior gonopods fused at the base, drawn 
anteriad as a knob articulating with both 
lateral sternal plates. Sternum of the an- 
terior gonopods divided, joined by a lightly 
sclerotized membrane; each sternal plate 
deeply cupped mesally, spiracle at lateral 
margin of cup, tracheal apodeme somewhat 
reduced. Posterior gonopods with the telop- 
odite artick^s subequal; colpocoxites with 
two branches, the anterior laminate, the 
posterior rodlike. Sternum of posterior 
gonopods much reduced, bandlike, deeply 
cui^ved posteriad between coxae; spiracle 
lateral, tracheal apodemes much reduced, 
bifurcate. 

DistriJ)ution. Upper Mississippi River 
Valley, presently known only from caves 
and mines. 

Species. One, the type species. 



Achemenides pecfinatus (Causey) 
Figures 10, 54-58 

Conottjla pectinata Causey, 1952, Proc. Biol. Soc. 

Wash., Vol. 65, pp. 112-113, figs. 4, 5. Cham- 

berlin and Hoffman, 1958, U. S. National Mus. 

Bull. 212, p. 99 (list). 
Sonorati/la pectinata Hoffman, 1961, Proc. Ent. 

Soc. Amer., Vol. 87, p. 268. 

Type. Male holotype from "Smith Park, 
Mt. Carroll, Carroll Co., Illinois"; in 
collection of Illinois Natural History 
Survey, examined. Mr. Stewart Peck, 
who has had access to the field notes 
of H. H. Ross and M. W. Sanderson, 
collectors of the holotype male, in- 
formed me that the type locality was 
actually Smith Park Cave. As the type 
consists of only a dozen legless mid- 
body segments, badly stained, and 
slide-mounted gonopods that could not 
be located, the description below is 
based on a specimen from the type 
locahty identified by Causey and 
placed in the Illinois Natural History 
Survey Collection. The specific epithet 
is a Latin adjective referring to the 
comblike appearance of the anterior 
gonopods. 
Description of male from Mt. Carroll, 
Illinois. Length, about 22 mm (specimen 
fragmented). Ocelli in three rows on each 
side, 14 in each triangular eyepatch. An- 
tennal segments in order of length: 3, 5, 4, 
2, 6, 7, 1. Legs 1 and 2 with slight swell- 
ings on the third joint, leg 4 with a weak 
apophysis (sometimes lacking) on the 
femur; legs 5 and 6 with strong distal 
apophyses on the femora; leg 7 with a 
capitate, toothed apophysis on the femur 
and a long, proximal lobe on the prefemur 
(Fig. 58). Anterior gonopods (Figs. 54, 55) 
very large, joined at base to form an an- 
teriorly projecting cur^'ed knob; posterior 
surfaces with a thin, deeply laciniated' 
ridge. Colpocoxites of posterior gonopods 
(Figs. 56, 57) with two divisions; an an- 
terior deeply laciniated lamella and a 
posterior sagittiform branch; first telop- 



MiLLiPED Family Conotylidae • Shear 



85 





0. 5 mm 



(^7^ ,c^ 




0.3 mm 
I ^ 






^^ 






62 1^ ,//,,, 





igures 56-64. Gonopods and associated structures of kchememdei and Taiyutyla. Figs. 56-58. Ac/iemen/des pecfinafus. 
ig. 56. Left posterior gonopod, posterior view. Fig. 57. Colpotoxite of rigfit posterior gonopod, anterior view. Fig. 58. 
igfit leg 7, posterior view. Figs. 59-60. Taiyutyla corvallis. Fig. 59. Anterior gonopods, posterior view. Fig. 60. Left 
losterior gonopods, posterior view. Figs. 61-62. T. francisca. Fig. 61. Anterior gonopods, posterior view. Fig. 62. Left pos- 
srior gonopod, posterior view. Figs. 63-64. 7. nopo. Fig. 63. Anterior gonopods, posterior view. Fig. 64. Posterior gono- 
ods, posterior view. 



86 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 2 



odite joint with a low, rugose swelling on 
posterior mesal side. 

Notes. I find the following diserepaneies 
between the original deseription and the 
topotype specimen: the apophysis of leg 5 
is on the fourth, not the third joint; leg 6 
has a single lolx> on the fourth joint, not 
two lobes on the third; leg 7 (Fig. 58) has 
apophyses on both the third and fourth 
joints, not just the third; there is no evi- 
dence of a "dorsal" (anterior) branch of 
the anterior gonopods; the deeply laciniate 
lamellae of the posterior gonopods are 
coxal, not sternal. 

The reduced ocelli and fairly large size 
of this species indicate that it is somewhat 
adapted for a subterranean existence. 
Anstrotijla specus, on the other hand, is 
troglophilic and shows no strong adapta- 
tions for cave life. Thus both could pos- 
sibly occur in Smith Park Cave without 
competing directly. The locality needs to 
be recollected to confirm this, and to pro- 
vide more specimens. The presence of this 
species at the edge of the classical Driftless 
Area, and in caves, is of great potential 
significance, especially since it shows a 
combination of conotyloid-austrotyloid fea- 
tures. But the problems associated with 
defining this area (Frye, 1965) limit specu- 
lation. Gushing (1965) found evidence of 
many disjunct arctic-alpine plants in the 
Driftless Area, and other plants that occur 
only south of the Wisconsin maximum 
farther east. It is not unlikely, therefore, 
that the region escaped glaciation, while 
the extreme climate drove A. pectinotiis 
or its ancestor to seek refuge in the moder- 
ated climate of caves. In view of this un- 
usual interest, I present a detailed list of 
new records below (see also Map 3). 

Records. ILLINOIS: Jo Daviess Co., 
mines in North California Diggings, 7 mi. 
NW of Hanover, 31 October 1965, S. Peck, 
$9 9; South Nicholsen Mine, 31 October 
1965, S. Peck, S $ 9; Hutchings Mine, 5 mi. 
E of Galena, 30 October 1965, S. Peck, $ 9 . 
IOWA: Jackson Co., Hunter's Cave, 5 mi. 
N of Andrew, January-February 1966, S. 



Peck, (5 5 5 5. WISCONSIN: Richland Co., 
John Gray Cave, 5.5 mi. NNE of Richland 
Center, no date, C. Krekc^ler, $ . 

Genus Taiyufyla Chamberlin, 1952 

Taiytityla Chamlieilin, 1952, Nat. Hist. Misc., \ 
Chicago Acad. Sci. No. 113, p. 1. Chamberlin 
and Hoffman, 1958, U. S. Nat. Mus. Bull. 212, 
p. 102. Hoffman, 1961, Trans. Amer. Ent. Soc, 
Vol. 87, p. 270. 

Type .species. Taiyutyla corvallis Cham- 
berlin, by original designation. The origin 
of the generic name, a neologism, is ob- 
scure. The gender is believed to be femi- 
nine. 

D/f/gno,s/.s. Small (S-11 mm in length) 
conotylids with the fifth antennal segment 
longer than the fourth, 20 ocelli or less in 
each eyepatch. Pregonopodal legs with 
apophyses on third or fourth article. 
Eleventh legs with prefemoral hooks. An- 
terior gonopods platelike, sometimes with 
terminal and subterminal processes or 
lamellae; stenium of anterior gonopods 
completely surrounding coxotelopodites, 
which articulate primarily on its posterior 
surface; spiracle prominent, in lateral de- 
pression. Posterior gonopods large, but 
slightly smaller to distinctly smaller than 
anterior gonopods, platelike or with two 
major branches, subterminal branch usually 
flattened, but sometimes spikelike, subtend- 
ing or bearing a plumose-hirsute area; 
terminal branch or process short, or long 
and spirally curved, often bearing accessory 
teeth; no coxal depression, coxal bases 
swollen posteriorly. Sternum of posterior 
gonopods broad and deep anteriorly, thin 
and ribbonlike posteriorly; spiracles at 
proximolateral margins of broad, semi- 
circular depressions. Telopodite articles of 
posterior gonopods \'ariable, in some cases 
subequal and with second showing signs of 
segmentation, or with the first twice as 
long as the second, which is reduced to a 
small faiob. 

Species and distrihution (Map 1). Three 
species in southern On^gon and northeni 
California. 



MiLLiPED Family Conot\'lidae • Shear 



87 



Key to Species 

la. Second telopodite joint of male posterior 
gonopods only half the length of the first 
(Fig. 64); leg 3 with an apophysis on the 
fourth joint, other pregonopodal legs un- 
modified napa 

lb. Second telopodite joint of male posterior 
gonopods (Figs. 60, 62) at least as long 
as the first; legs 4 through 7 and sometimes 
leg 3 with femoral apophyses 2 

2a. Third leg of male with an apophysis on the 
femur; gonopods as in Figs. 61, 62; Marin 
Co., California francisca 

2b. Third leg of male unmodified; southern 
Oregon corvallis 

Taiyutyla corvallis Chamberlin 
Figures 8, 59, 60 

TaiyiityJa corvallis Chamberlin, 1952, Nat. Hist. 
Misc., Chicago Acad. Sci. No. 113, pp. 1-2, 
figs. 1, 2. Chamberlin and Hoffman, 1958, U. 
S. Nat. Mus. Bull. 212, p. 102 (list). Hoffman, 
1961, Trans. Amer. Ent. Soc, Vol. 87, p. 270, 
pi. 10, figs. 8, 9. 

Type. Holotype from Corvallis, Linn 
Co., Oregon; in Chamberlin collection, 
Salt Lake City, Utah, could not be 
located. 

Diagnosis. Unique in lacking an apoph- 
ysis on the fourth joint of leg 3. 

Description of mole from Corvallis, Ore- 
gon. Length, 10.5 mm. Eyepatches tri- 
angular, three rows plus single ocellus, 
totaling 17, on right side; 20 ocelli in four 
rows on left side. Antennal articles in order 
of length: .3, 5, 4, 2, 6 = 7, 1. Legs 4 
through 7 with apophyses on femora; 
apophyses on legs 4 and 5 veiy small, 
distal; apophysis of leg 6 strong, capitate, 
mesal; apophysis of leg 7 strong, capitate 
and toothed, promixal. Anterior gonopods 
( Fig. 59 ) larger than colpocoxites of poste- 
rior gonopods, subrectangular, flattened 
antero-posteriorly, slightly bent mesad, 
with aciculate subterminal process arising 
From a longitudinal ridge; lateral edges 
raguely serrate. Sternum of anterior gono- 
pods completely encircling coxotelopodites, 
produced between them as bilobed or 
trilobed condyle. Colpocoxites of posterior 
gonopods (Fig. 60) large and bulbous at 



base, ending in pointed terminal process, 
below which is a pilose area and a pointed 
subterminal process. Second joint of pos- 
terior gonopod with a distinct basal 
shoulder; third joint nearly twice as long 
as second. 

Notes. Nothing is known of the ecology 
and biology of this species, which is known 
only from the type locality (Map 6). 

Taiyutyla napa n. sp. 
Figures 63, 64 

Type. Holotype male collected with two 
females and an immature specimen by 
Vincent Roth, 31 December 1953, Mt. 
St. Helena, Napa and Sonoma cos., 
California. The specific epithet refers 
to the type locality. 

Diagnosis. In both corvallis and fran- 
cisca, the second telopodite article of the 
posterior gonopods is longer than or as 
long as the first; in napa the second joint 
is less than half as long as the first. Only 
leg 3 has an apophysis on the fourth joint 
in napa; corvallis has leg 3 unmodified and 
francisca has apophyses on legs 4 through 
7, as well as 3. The low swelling on the 
third joint of leg 2 present in napa is absent 
in the other two species. 

Description of liolotype male. Length, 8 
mm. Eyepatches irregularly triangular, 3 
irregular rows with a single ocelkis totalling 
18 on right side; IS ocelli in four rows on 
left side. Antennal joints in order of length: 
3, 5, 4, 2 = 6, 7, 1; joint 5 about twice as 
long as joint 4. Leg 3 with large distal 
apophysis on femur; leg 2 with a low, 
longitudinal swelling on the mesal side of 
third joint. Anterior gonopods (Fig. 63) 
curv^ed posteriad distally, with prominent 
mesal shoulder about midway in their 
length, and with a thin, membranous sub- 
terminal lamina extending mesally. Colpo- 
coxites of posterior gonopods (Fig. 64) 
intermediate between corvallis and fran- 
cisca, subterminal process short, pointed; 
terminal process long, curved posteriad, 
with single accessory tooth. First telop- 



88 



BuUetin Museum of Comparative Zoology, Vol. 141, No. 2 






Map 6. Northern California and southern Oregon, showing distribution of various conotylids; dots, Plumatyla humerosa. 
circle, Conotylo extorns; solid squares, Ta/yutylo francsco; open squares, 7. napa,- triangles, T. corvol/is, including 7. 

jonesi. 

Map 7. Southern Indiana, showing distribution of Conofylo bo/lmoni; dots, epigean records; circles, cave records; dottec 

line, limit of lllinoisan glacial drift; broken line with dots, limit of Wisconsin glacial drift. 



odite joint more than twice as long as 
second. 

Notes. Nothing is known of the biology 
of this species, which is known only from 
the type locality (Map 6). Mt. St. Helena 
has a maximum elevation of 4344 feet. 

Taiyufyla francisca n. sp. 
Figures 61, 62 

Type. Holotype male collected by C. W. 
O'Brien, 7 Januaiy 1962, one mile SE 
of Inverness, Marin Co., California. 
The specific epithet, a noun in appo- 
sition, refers to the proximity of the 
type locality to San Francisco Bay. 



Diaii,nosis. Distinct from corvallis pri- 
marily in the form of the colpoeoxites o\ 
the posterior gonopods, which bear spirally 
curved apical processes with several acces- 
sory teeth, also in the stronger modifi- 
cations of legs 4 and 5 of the males. 

Description of liolotype mole. Length 
9.5 mm. Eyepatches triangular, 20 ocell: 
in four ro\\'s on left side, 20 ocelli in foui 
rows and single ocellus on right side. An- 
tennal segments in order of length: 3, 5, 4. 
2, 6, 7, 1. Apophyses of legs 3 through "i 
approximately same size, distal on leg 3 
slowly becoming mesal on succeeding leg? 
and proximal on leg 7. Anterior gonopodf 



MiLLiPED Family Conotylidae • Shear 



89 



(Fig. 61) large, flattened, terminal process 
blunt, slightly curved; large, thin mesal 
flange present. Colpocoxites of posterior 
gonopods (Fig. 62) smaller than anterior 
gonopods, complex; terminal process curved 
in a spiral of two turns, with four accessory 
processes as shown (Fig. 62); subterminal 
process a twisted, horizontal lamella finely 
laciniate on the posterior side. Second 
telopodite joint of posterior gonopod 
slightly longer than first, vaguely annulated. 
Notes. Little is known of the biology of 
this species; the holotype was taken in a 
Berlese sample of Piniis mtiricata duff. See 
Map 6. 

Genus Austrotylo Causey, 1961 

Austwtyla Causey, 1961, Proc. Biol. Soc. Washing- 
ton, Vol. 74, p. 260 (in part). 

Soiwiati/la Hoffman, 1961, Trans. Ent. Soc. Amer., 
Vol. 87, p. 269. 

Type species. Of Austrotyhu ConotyJa 
specus Loomis, by original designation; of 
Sonoratyhi, ConotyJa montivaga Loomis. 
The generic name is a feminine Latin- 
Greek neologism referring to the southerly 
distribution of the genus with respect to 
Conotyla. 

Diapiosis. With the characters of the 
family. Anterior gonopod sternum appear- 
ing divided in some cases, but usually 
contiguous in the anterior and posterior 
midlines, or joined by sclerotic membrane. 
Anterior gonopods flattened, platelike, but 
with complex posterior surfaces; anterior 
gonopods covering colpocoxites of posterior 
gonopods in situ. Colpocoxites of posterior 
gonopods much smaller than anterior gono- 
pods, usually with a single cupped lamella 
and a rodlike or platelike mesal branch, 
sometimes plumose. Pregonopodal legs 
with femoral lobes on legs 3 and 4 in all 
known species. Capitate lobes present on 
coxae of legs 10 and 11 in some species. 
Species usually pigmented, 20-24 ocelli in 
triangular j)atch. 

Species. Five; distributed (Map 1) 
through Missouri and Illinois, Rocky Moun- 
tains from Alberta to Chihuahua (and 



possibly Queretaro). Separation of specus 
or its ancestors from the Rocky Mountain 
species may have occurred as early as the 
Pliocene, at the time of the formation of 
the Great Plains. 

Key to Species 

la. Coxae 10 and 11 of males without lobes .— 

horealis 

lb. Coxae of legs 10 and 11 with lobes, or 

either coxa 10 or 11 lobed 2 

2a. Lobe on coxa 11 only ._ coloradensis 

2b. Lobe on coxa 10 3 

3a. Lobe on coxa 10 only 4 

3b. Lobes on both coxae 10 and 11 specus 

4a. Gonopods as in Figs. 68-70 rnontivaga 

4b. Gonopods as in Figs. 77-79 chihuahua 

Austrotyla specus (Loomis) 
Figures 65-67 

Conotyla specus Loomis, 1939, Bull. Mus. Comp. 

Zool., Vol. 86, p. 184, figs, lla-c. Chambedin 

and Hoffman, 1958, U. S. Nat. Mus. Bull. 212, 

p. 99 (list). 
Ausfrotyhis specus specus, Causey, 1961, Proc. 

Biol. Soc. Wash., Vol. 74, pp. 260-264, figs. 

5-10. 
Austrotyla specus rnontivaga, (in part) Causey, 

196.1, Proc. Biol. Soc. Wash., Vol. 74, pp. 264- 

265. 
Sonoratyla specus, Hoffman, 1961, Trans. Ent. 

Soc. Amer., Vol. 87, p. 269. 

Type. Male holotype from Rice's Cave, 
3 miles northeast of Goldman, Jeff- 
erson Co., Missouri, in Museum of 
Comparative Zoology, examined. The 
specific epithet is a noun in apposition 
(Latin: "cave") referring to the habi- 
tat of the type series. 
Diagnosis. This species has much less 
complex anterior gonopods than colora- 
densis or chihuahua, and those of horealis 
are much simpler. In addition, the coxal 
processes of legs 10 and 11 are unique. 
See under rnontivaga for a discussion of 
differences between that species and 
specus. 

Description of holotype male. Length, 
11.5 mm. Eyepatches quadrangular, ocelli 
of both sides in three rows, 21 on the left 
side, 23 on the right, ocelli fully pigmented, 
not irregular in shape. Antennal articles in 



90 Bulletin Museum of Contparativc Zoology, Vol. 141, No. 2 




0. 3 mm 






Figures 65-73. Gonopods of Austrotyla. Figs. 65-67. Austrotyla specus. Fig. 65. Right posterior gonopod, anterior view. 
Fig. 66. Anterior gonopods, posterior view. Fig. 67. Left anterior gonopod, lateral view. Figs. 68-70. A. montivaga. Fig. 
68. Right posterior gonopod, anterior view. Fig. 69. Anterior gonopods, posterior view. Fig. 70. Left anterior gonopod, 
lateral view. Figs. 71-73. A. borealis. Fig. 71. Right posterior gonopod, anterior view. Fig. 72. Anterior gonopods, posterior 
view. Fig. 73. Left anterior gonopod, lateral view. 



MiLLiPED Family Conotylidae • Sheai 



91 



order of length: 3, 5, 4, 6, 2, 7, 1. Pre- 
gonopodal legs modified as described for 
the genus, legs 10 and 11 with anteriorly 
directed capitate lobes on the coxae. An- 
terior gonopods (Figs. 66, 67) with the 
sternal lobes evenly and heavily sclerotized, 
coxotelopodites with a lateral lobe above 
the sternal lobes; proximal laciniated 
branches large, prominent. Colpocoxites 
( Fig. 65 ) with the posterior lamella cupped 
and rounded distally, anterior branch sig- 
moid, laciniated distally; third telopodite 
joint twice the length of the second. 

Notes. Both Causey and Loomis over- 
looked the lobes on the coxae of the Uth 
legs. Causey noted the variation in color- 
ation and ocelli of this species, and a study 
of other populations indicates that both 
pigmented and unpigmented individuals 
occur in the same caves. Epigean speci- 
mens are rare; Causey reported them as 
A. s. montivaga from northern Illinois and 
Wisconsin. It seems likely that the same 
situation holds in this species as in Conotyla 
blakei, with troglophilic populations in the 
southern, lowland part of the range and 
epigean populations in the northern part. 
The absence of epigean records of either 
species in the regions of the cave popu- 
lations may be due to a lack of collecting 
at the proper time, late fall, winter, and 
early spring, when these animals mature 
and are most active. Unpublished records 
kindly given to me by Stewart Peck include 
:-aves in Jackson Co., Iowa, and in the 
Following Illinois counties: Jackson, Jo 
Daviess, Monroe, Saline, Henderson, and 
Union. Causey (1961) reported it from 
Franklin, Jefferson, and St. Clair cos., 
Missouri, and Sauk Co., Wisconsin (Map 
5). An immature specimen in the Museum 
jf Comparative Zoology from Blue Earth 
oo., Minnesota, strongly resembles A. 
>pecus. 

^usfrotyla montivaga (Loomis) 
Figures 68-70 

'Conotyla montivaga Loomis, 1943, Bull. Mus. 
Comp. Zool., Vol. 92, pp. 383-384, figs. 4a-d. 



Chambeilin and Hoffman, 1958, U. S. Nat. 

Mus. Bull. 212, p. 98 (list). 
Sonoratyla montwap,a, Hoffman, 1961, Trans. Ent. 

Soc. Amer., Vol. 87, p. 268, pi. 10, figs. 10-11. 
Austrotijla specus montivaga, (in part) Causey, 

1961, Proc. Biol. Soc. Washington, Vol. 74, p. 

264. 

Type. Male holotype from Santa Rita 
Mts., elevation 7500 ft., Pima Co., 
Arizona, in Museum of Comparative 
Zoology, examined. 
Diagnosis. This species is much smaller 
than coloradensis, and the posterior surface 
of the anterior gonopods is much less com- 
plex; there is no clear area on the lateral 
sternal lobes as there is in coloradensis and 
chihuahua. Distinct from specus, which it 
closely resembles, in the details of the gono- 
pods; in specimens I have examined, the 
third telopodite article is nearly three times 
as long as the second in montivaga, while 
in specus it is only slightly more than twice 
as long; there is no coxal lobe on leg 11 in 
montivaga, and one is present in specus. 

Description of liolotype male. Length, 
9.0 mm. Causey ( 1961 ) described in detail 
the nonsexual characters of this species, but 
stated that the gonopods (Figs. 68, 69) 
were identical to those of specus. This is 
not the case. In posterior view (Fig. 69), 
the proximal laciniated lobes of montivaga 
are smaller, the ridges and lobes are less 
developed, and there is no lateral extension 
on the coxotelopodite above the insertion 
of the sternal lobes. The colpocoxites ( Fig. 
68) are less developed in montivaga, and 
the anterior branch is thicker and blunter. 
Otherwise, the description given by Causey 
(1961) is accurate. 

Notes. Known from numerous specimens 
collected in the Santa Rita and Santa Cata- 
lina Mts., north and south of Tucson, Pima 
Co., Arizona. The type is from an elevation 
of 7500 ft., probably in or just below the 
Pinus ponderosa zone at the latitude of 
Tucson. The identity of specimens from 
Mescalero, New Mexico, could not be 
checked, as they were not available for 
study, but Chamberlin reported colora- 



92 BuUctin Museum of Comparative Zoo/ogry, Vol 141, No. 2 



densis from Riiidosa, in the same mountains 
and onl\' about 20 mil(\s to the northeast. 
^^"hie]l spi^eies (or an undescribed one) 
aetually oecurs in soutliem New Mexico 
awaits clarification. Both Loomis and 
Chamberhn mi<2;ht have assumed tliat their 
species was the only one in the Rocky 
Mountains; coJorodensis males were not 
described until 1961. The records reported 
by Causey ( 1961 ) from Illinois and Wis- 
consin refer to epigean populations of 
specus; she stated that the gonopods were 
identical with speciis, and placed monti- 
V(i<ia as a subspecies of speciis. A compari- 
son of my figures of the holotypes of both 
species should establish their distinctness, 
besides the geographic difficulties of hav- 
ing a single subspecies with two popu- 
lations separated by nearly a thousand 
miles of uninhabitable terrain. 

Ausfrofyla borealis n. sp. 
Figures 71-73 

Ti/pe. Male holotvpe collected by D. 
Whiteh(>a(l 2-4 October, 1967, Jasper 
National Park, Alberta, "Sta. 5," 5300 ft. 
The specific epithet is an adjective indicat- 
ing that it is the extreme northerly repre- 
sentative of its genus. 

Diagnosis. This species is smaller than 
the other representatives of the genus, 
being about 9 mm long in mature males. 
The stenia of the anterior gonopods show 
very distinct angular shoulders while still 
attached to the telopodites, as in specus. 
The posterior surface of the coxotelopodites 
is simpler than in any other species, and 
there are no lobes on the coxae of legs 10 
and 11. 

Description of holotijpe nude. Length, 9.S 
mm. Eyepatches triangular; 24 ocelli on 
left side, in four rows plus single ocellus; 
four ro\\s plus single ocellus totalling 22 
ocelli on right side. Antennal segments in 



order of length: 3, 5, 4, 6, 2, 7, 1. Legs 3 
and 4 modified as described for genus. 
Lateral lobes of the sterna of the anterior 
gonopods forming a distinct angular 
shoulder with the telopodites (Figs. 72, 
73), with a distal area of very thin cuticle; 
spiracle on anterior surface, easily seen. 
Coxotelopodites of anterior gonopods 
simple, with single thickened ridge on 
posterior surface, ending in somewhat 
swollen knob. Colpocoxites of posterior 
gonopods (Fig. 71) with anterior branch 
long, filiform, branched. First tc^lopodite 
joint is less than one third length of second 
telopodite joint. Coxae 10 and 11 without 
lobes. 

Notes. Known only from the type lo- 
cality. See notes on ConotyJa olhertana for 
further details on type locality. 

Ausfrofyla coloradensis (Chamberlin) 
Figures 74—76 

Conott/la coloradensis Chamberlin, 1910, Ann. Ent. 

Soc." Amer., Vol. 3, p. 237, pi. 32, figs. 7-9, pi. 

33, figs. 1-3. Chamberlin and Hoffman, 1958, 

U. S. Nat. Mus. Bull. 212, p. 98 (list). 
Austrotijla coloradensis. Causey, 1961, Proc. Biol. 

Soc. Washington, Vol. 74, pp. 254-260, figs. 

2-4. 

Type locolity. Colorado. Causey (1961) 
designated a male neotype from Allen's 
Park, Boulder Co., Colorado, which is 
deposited in the Museum of Compar- 
ative Zoology, examined. 
Diagnosis. Distinct in size and com- 
plexity of gonopods from all except duJiita- 
hua, but coloradensis has no anterior lobe 
on the coxa of leg 10. Distinct from monti- 
vaga by the larger size and much more 
prominent lateral shoulders. 

Description of neotype male. Length, 
15.7 mm. Eyepatches triangular, 23 ocelli 
on each side in four rows plus single 
ocellus. Antennal articles in order of length: 
3, 5, 4, 6, 2, 7, 1. Pregonopodal legs modi-' 



Figures 74-81. Gonopods of Austrotyla and Plumatyla. Figs. 74-76. Austrotyla coloradensis. Fig. 74. Right posterior gono- 
pod, anterior view. Fig. 75. Anterior gonopods, posterior view. Fig. 76. Left anterior gonopod, lateral view. Figs. 77-79. 



MiLLiPED Family Conotylidae • Shear 93 





0.3 mm 




^- chihuahua. Fig. 77. Colpocoxite of righf posterior gonopod, anterior view. Fig. 78. Anterior gonopods, posterior view, 
■ig- 79. Left anterior gonopod, lateral view. Figs. 80-81. Plumatyla humerosa. Fig. 80. Right anterior gonopod, anterior 
lew. Fig. 81. Left posterior gonopod, posterior view. 



94 Bulletin Museum of Comparative Zoology, Vol. 141, No. 2 



fied as described for genus, apophysis of 
leg 3 sonie\\'hat larger than that of leg 4. 
Anterior gonopods (Figs. 75, 76) thin, 
lamellate, bearing posteriorly a prominent 
knob near the midline of each telopodite 
and a mesal laciniated branch; sternal lobes 
with a semicircular area of very thin 
cuticle that appears to be a hole at low 
magnification. Colpocoxites of the posterior 
gonopods (Fig. 74) relatively large, deeply 
cupped, posterior lamella deeply notched 
laterally, anterior branch small, lightly laci- 
niated; second telopodite joint twice the 
length of first. 

Notes. Causey's (1961a) drawings of 
the gonopods, made from cleared, slide- 
mounted material, leave something to be 
desired. The thin area on the sternal lobes 
is represented as an open space, and the 
tracheal apodeme is shown as being 
coalesced with the sternal lobe high up 
on the coxotelopodite. In actuality, the 
tracheal spiracle, not seen by Causey, is 
easily visible on uncleared preparations and 
is at the lateroanterior comer of the 
stemiun, whence it leads normally into the 
tracheal apodeme. The horizontal portions 
of the sternites meet in the anterior midline 
as well as the posterior, rather than being 
represented entirely by a membranous area. 
The colpocoxites of the posterior gonopods 
in Causey's illustration show neither the 
deep lateral cleft in the posterior lamella, 
nor the short, mesal plumose branch. See 
Causey (1961a) for emendations in Cham- 
berlin's original description, based on a 
female. 

There is some variability in size in this 
species, but it could not be connected with 
any geographical trend. The smallest speci- 
mens were about 16 mm long and the 
longest were close to 23 mm long. 

KnowTi from numerous specimens from 
the following Colorado counties (Map 4): 
Larimer, Jackson, Eagle, Pitkin, Chaffee, 
Gunnison, Hinsdale, Mineral, and Conejos. 
Probably also occurs in southern Wyoming 
and northern New Mexico. The majority of 



records are from coniferous forests above 
7000 ft. elevation. 

Austrotyla chihuahua n. sp. 
Figures 77-79 

Type. Male holotype from 100 m above 
Rio Urique, 84 km south of Creel, 
Chihuahua, Mexico, collected Febru- 
ary 28, 1966, by J. Reddell and W. 
Bell. The specific epithet is a noun in 
apposition, referring to the type lo- 
cality. 
Diaf^nosis. The posterior surface of the 
anterior gonopods is complex, as in colora- 
dc'nsis\ but the colpocoxites of the posterior 
gonopods of chihuahua have the posterior 
lamella angular and the anterior branch 
large, flattened and reflexed; chihuahua 
has a coxal lobe on leg 10, while colora- 
densi.s has none. 

Description of holotype male. Length, 
14.0 mm. Eyepatches tiuncate-triangular, 
23 ocelli in four rows on left side, 21 ocelli 
in four rows on right side. Antennal articles 
in order of length: 3, 5, 4, 2 = 6, 7, 1. 
Legs 3 and 4 modified as described for 
genus, apophyses slightly larger than in 
other species. Sterna of anterior gonopods 
as in coloradensis; gonopods (Figs. 78. 79) 
almost contiguous in midline, distinctly 
depressed mesally on anterior surface; 
posterior surface with complex knobs and 
plumose branches. Colpocoxites of poste- 
rior gonopods (Fig. 77) with posterior 
lamella triangular; anterior branch flat- 
tened, reflexed. Coxa of leg 10 \\'ith an 
anterior lobe. 

Notes. Nothing is known of the biology 
of this species, which is known only from 
the type locality. 

Genus Plumafyla, new genus 

Austrotijla Causey, (in part) 1961, Pioc. Biol. Soc, 

Washington, Vol. 74, p. 260. 
Sonorattjla Hoffman (in part) 1961, Trans. Amer 

Ent. Soc, Vol. 87, p. 269. 

Type species. Conotyla humerosc 
Loomis; the generic name is a Latinized 



MiLLiPED Family Conotylidae • Shear 95 



Spanish-Greek neologism derived from the 
related genus Conotyla and Plumas Co., 
California, type localit)- of the type species, 
rhe gender is feminine. 

Diagnosis. With the characters of the 
Family. Anterior gonopod sternum inter- 
mediate between Toiyutyla and Austrotyla, 
leavily sclerotic throughout, with lateral 
obes extending laterad to coxotclopodites, 
out incomplete posteriorly, as in Cono- 
"ylo. Anterior gonopods with two major 
tranches, the anterior largest and set 
nesally with small laciniate processes; 
posterior branch a simple hirsute rod. 
Posterior gonopod sternum broadened and 
lepressed laterally. Colpocoxites of poste- 
•ior gonopods with a large mesal branch 
)earing laciniations as in anterior gonopod; 
ateroposterior lamella heavily sclerotized. 
^emoral lobes on some pregonopodal legs. 
Jpecies troglobitic, without pigment, ocelli 
ibout 10, in two rows. 

Species. One, found in mines and caves 
n northern California. 

'/umofy/a humerosa (Loomis) 
Figures 80, 81 

lonotijla humerosa Loomis, 1943, Bull. Mus. 
Comp. Zool., Vol. 92, pp. ,384-385, figs. 5a-cl. 

Type. iMale holot\'pe and other speci- 
mens collected Sunnyside Mine, 3 mi. 
SW of Seneca. Plumas Co., California, 
January 22, 1923, by H. S. Barber; 
deposited in U. S. National Museum, 
examined; immature male paratype in 
Museum of Comparative Zoology. 
Diagnosis. See generic diagnosis. 
Description of topotype male. Length, 
6.0 mm. Ocelli in two rows, 9 ocelli on 
?ft side, 8 ocelli on right side. Antennal 
3gments in order of length: 3, 5, 4, 2, 6, 
, 1. Legs 5, 6, and 7 with prominent 
smoral lobes. Anterior gonopods (Fig. 
0) larger than colpocoxites; posterior 
ranch bears a subterminal lateral hook 
nd is distally bifid; posterior branch rod- 
ke, densely pilose. Posterior basal knob 
ts into sternal cavity of posterior gono- 



pods. Colpocoxites of posterior gonopods 
(Fig. 81) with two branches, anterior 
mesal branch hooked posteriad, small 
laciniated branches on lateral surface; poste- 
rior lateral branch a subtriangular, curved, 
well-sclerotized lamella. Tracheal apodemes 
of both gonopods reduced in size. 

Notes. The only other mature specimens 
known to me, excepting the specimens 
from the type locality kindly lent to me 
by N. B. Causey, are from Indian Wells Ice 
Cave, Lava Beds National Monument, 
Siskiyou Co., California. They differ from 
the holotype and paratype by being smaller 
(13-14 mm), lacking a lobe on the femur 
of leg 5, and ha\dng the anterior branch of 
the anterior gonopod slightly more attenu- 
ate. Otherwise, the gonopod structure is 
identical to the Sunnyside Mine specimens. 
For this reason, I hesitate to describe it as 
a distinct species. Immature representatives 
of Plumatyla are known from a number of 
mines, limestone caves, and lava tubes in 
northern California and adjacent Oregon. 
Only the collection of mature specimens 
can iudicate the range of variation and the 
number of species in this genus. The lava 
tubes in Lava Beds National Monument 
may be as old as 60,000 years (Gale, 1959). 
The local glaciation of the northern Cali- 
fornia area during the late Pleistocene 
(Detling, 1968) may have been responsible 
for the cave habitat of these animals. 

LITERATURE CITED 

Barr, T. C. 1968. Cave ecology and the evo- 
lution of troglohites. Evol. Biol., 2: 35-102. 

Broleaiann, H. W. 1935. Myriapodes Diplopodes 
( Chilognatlies I) Faiiue dc France, 29, Paris. 
368 pp. 

Buckett, J. S., AND M. R. Gardner. 1967. A 
new family of cavemicolous millipedes with 
tlie description of a new genus and species 
from Idaho (Diplopoda: Chordeumida: Chor- 
deumidea). Michigan Ent., 1(4): 117-126. 

Causey, N. G. 1952. Four new chordeumid 
millipeds from the United States (Nema- 
tophora, Chordeumidea). Proc. Biol. Soc. 
Washington, 65: 111-118. 

. 1961a. Austrotyla, a new milliped genus 

( Chordeumidea : Conot>'lidae : Conotylinae ) . 
Proc. Biol. Soc. Washington, 74: 251-266. 



96 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 2 



. 19611). Adrityh, a new millipecl genus 

(Chordeuinidea: Conotylidae). Psyche, 68: 

131-136. 
Chamberlin, R. V. 1941. New western millipeds. 

Bull. Univ. Utah, Biol. Sen, 6(5): 3-23. 
. 1951. Eleven new western millipeds. 

Natur. Hist. Misc. No. 87: 1-12. 

1952. Two Oregon millipeds of the Order 



Chordeumida. Nat. Hist. Misc., No. 113: 

1-3. 
Chamberlin-, R. V., Axn R. L. Hoffman. 1958. 

Checklist of the millipeds of North America. 

U. S. Nat. Mus. BuU., 212: 1-236. 
Cook, O. F. 1896. On recent diplopod names. 

Brandtia, 2: 8. 
Cook, O. F., and G. N. Collins. 1895. The 

Craspedosomatidae of North America. Ann. 

New York Acad. Sci., 9(1): 1-100. 
CusHiNG, E. J. 1965. Problems in the Quaternary 

phvtogeography of the Great Lakes region. 

/h'h. E. Wright, Jr. and D. G. Frey [Ed.] 

The Quaternary of tlie United States. Prince- 
ton, N. T-: Princeton University Press, pp. 

403-417. " 
Detling, L. E. 1968. Historical background of 

the flora of the Pacific Northwest. Bull. Mus. 

Nat. Hist. Univ. Oregon, 13: 1-57. 

FrYE, J. C, H. B. WiLLMAN, AND R. F. BlACK. 

1965. Outline of the glacial geology of Illi- 
nois and Wisconsin. In H. E. Wright, Jr. 
and D. G. Frey [Ed.] The Quaternary of the 
United States. Princeton, N. J.: Princeton 
University Press, pp. 43-61. 

Gale, R. R. 1959. Geology of Lava Beds National 
Monument. Bull. Nat. Speleol. Soc, 21(2): 
62. 

Hoffman, R. L. 1961. Systematic and morpho- 
logical notes on North American conotyloid 



Diplopoda. Trans. Amer. Ent. Soc., 87: 259- 

272. 
Hoffman, R. L. 1963. Notes on the structure 

and classification of the diplopod family , 

Heterochordeumatidae. Ann. Mag. Nat. Hist., ' 

13(6): 129-135. 
Howden, H. F. 1963. Speculations on some 

beetles, barriers, and chmates during tlie 

Pleistocene and Pre-Pleistocene periods in 

some non-glaciated portions of North America. 

Syst. Zool., 12(4): 178-201. 
LooMis, H. F. 1939. The millipeds collected in 

Appalachian caves by Mr. Kenneth Dearolf. 

Bull. Mus. Comp. Zool., 86(4): 165-193. 
. 1943. New cave and epigean millipeds. 

Bull. Mus. Comp. Zool., 92(7): 373-410. 
Moss, E. H. 1955. Vegetation of Alberta. Bot. 

Rev., 21: 493-567. 
ScHUBART, O. 1934. Tausendfiissler oder Myriap- 

oda. 1: Diplopoda. In Die Tierwelt Deutsch- 

lands, 28 Teil. Jena: Gustav Fischer, 318 pp 
Shear, W. A. 1969. Cave miUiped genera of tht 

United States. Psyche, 76(2): 126-143. 
Verhoeff, K. 1913. Ascospermophoren aus Japan 

Zool. Anz., 43: 342-370. 
. 1932. Diplopoden-Beitrage (124. Diplo 

poden-aufsatz ) . Zool. Jahrb., Syst. Arb., 62: 

469-524. 

1942. Ascospermophoren aus Japan unc 



iiber neue japanische Diplopoden. Zool. Anz. 
137: 201-217. 
Wayne, W. J., and J. H. Zumherge. 1965. Pleisto 
cene geology of Indiana and Michigan, h 
H. E. Wright, Jr. and D. G. Frey [Ed.] Th. 
Quaternary of the United States. Princeton- 
N. J. Princeton Univ. Press, pp. 63-83. 

[Received 10 October 1969.) 



MiLLiPED Family Conotylidae • Shear 



97 



Valid generic and species names in italics; 

Achemenides, 84 
Adritylidae, 57-58 
aeto, Conotyhi, 76 
albeitana, Conotyla, 82 
utrolineata, Conotyla, 80 
\twliucata group of Conotyla, 80 
Atistrotyla, 89 

hlakei, Conotyla, 67 
Blakei group of Conotyla, 67 
blakei, Proconotyla, 67 
boUmani, Conotyla, 69 
oollmani, Trichopetalum, 69 
horealis, Atistrotyla, 92 

zeleno, Conotyla, 80 
:hihiiahua, Austrotyla, 94 
Uleidogonidae, 56 
:oloraclen.sis, Austrotyla, 92 
:oloradensis, Conotyla, 92 
'Jonotyla, 64-67 
Donotylidae 

description, 58 

biology, 58-60 

gonopod structure, 61-62 

key to genera, 64 

in key to related families, 57 

problematical names, 62-64 

taxonomic position, 56 
^orvallis, Taiytttyla, 87 

Diplomaragnidae, 56 

dpcnor, Conotyla, 73 
Zudifiona, 56 
uuligoninae, 56 
'xtorris, Conotyla, 70 

ischeri, Conotyla, 72 

uscheri group of Conotyla, 72 

rancisca, Taiyutyla, 88 

^lomeratum, Trichopetalum, 63 



INDEX 

only major discussions listed for species. 

humerosa, Phimatyla, 95 
humerosa, Conotyla, 95 

Idagonidae, 56, 57 

Japanosoma, 56 
Japanosomatinae, 57 
jonesi, Conotyla, 63 

leibergi, Cookella, 63 

Marquetia ( Opisthocheiridae ) , 57 
mclinda, Conotyla, 77 
Melinda group of Conotyla, 77 
montivaga, Austrotyla, 91 
montivaga, Conotyla, 91 
montivaga, Sonoratyla, 91 

napa, Taiyutyla, 87 

ocypetcs, Conotyla, 76 
oregona, Bollinanclla, 63 

pectinatus, Achemenides, 84 
pectinata, Conotyla, 84 
pectinata, Sonoratyla, 84 
personata, Conotyla, 73 
pliana, Zygotyla, 63 
Plumatyla, 95 
Proconotyla, 64 

smilax, Conotyla, 75 

Smilax group of Conotyla, 75 

Sonoratyla, 89, 94 

specus, Austrotyla, 89 

specus, Conotyla, 89 

specus, Sonoratyla, 89 

Taiyutyla, 86 
Trichopetalidae, 55, 56, 57 

vaga, Conotyla, 67 
venetia, Conotyla, 75 
Venetia group of Conotyla, 75 
visia, Conotyla, 79 

Wyandotte, Conotyla, 69 
Wyandotte, Scotherpes, 69 



mwmM^^Mm^Wh^^:^v'/> : 



\ • 



OF THE 



Museum of 

Comparative 

Zoology 



Monograph of the Cuban Genera 

Emoda and Glyptemoda 

(Mollusca: Archaeogastropoda: Helicinidae] 



WILLIAM J. CLENCH AND MORRIS K. JACOBSON 



HARVARD UNIVERSITY VOLUME 141, NUMBER 3 

CAMBRIDGE, MASSACHUSETTS, U.S.A. 4 FEBRUARY 1971 



PUBLICATIONS ISSUED 

OR DISTRIBUTED BY THE 

MUSEUM OF COMPARATIVE ZOOLOGY 

HARVARD UNIVERSITY 

Bulletin 1863- 

Breviora 1952- 

Memoirs 1864-1938 

JoHNSONiA, Department of Mollusks, 1941- 

OccAsiONAL Papers on Mollusks, 1945- 



Other Publications. 

Bigelow, H. B., and W. C. Schroeder, 1953. Fishes of the Gulf of Maine. 
Reprint, $6.50 cloth. 

Brues, C. T., A. L. Melander, and F. M. Carpenter, 1954. Classification of 
Insects. $9.00 cloth. 

Creighton, W. S., 1950. The Ants of North America. Reprint, $10.00 cloth. 

Lyman, C. P., and A. R. Dawe (eds.), 1960. Symposium on Natural Mam- 
malian Hibernation. $3.00 paper, $4.50 cloth. 

Peters' Check-list of Birds of the World, vols. 2-7, 9, 10, 12-15. (Price list 
on request. ) 

Turner, R. D., 1966. A Survey and Illustrated Catalogue of the Teredinidae 
( MoUusca: Bivalvia ) . $8.00 cloth. 

Whittington, H. B., and W. D. I. Rolfe (eds.), 1963. Phylogeny and Evolution 
of Crustacea. $6.75 cloth. 

Proceedings of the New England Zoological Club 1899-1948. ( Complete sets 
only. ) 

Publications of the Boston Society of Natural History. 



Authors preparing manuscripts for the Bulletin of the Museum of Comparative 
Zoology or Breviora should send for the current Information and Instruction Sheet, 
available from Mrs. Penelope Lasnik, Editor, Publications Office, Museum of 
Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138, 
U.S.A. 



© The President and Fellows of Harvard College 1971. 



MONOGRAPH OF THE CUBAN GENERA EMODA AND GLYPTEMODA 
(MOLLUSCA: ARCHAEOGASTROPODA: HELICINIDAE) 



WILLIAM J. CLENCH AND MORRIS K. JACOBSON 



TABLE OF CONTENTS 

Introduction 99 

Acknowledgments 101 

Genus Einoda H. and A. Adams 101 

Key to Species 103 

Emoda siJacea (Morelet) 104 

£. sa^iaiana (Orbigny) 105 

E. pulclicrrima pidcherrima (Lea) 107 

E. ))ulchcrrima titanica (Poey) 108 

E. suhmarginata (Gray) 109 

E. ciliata (Poey) 110 

E. cilUita f^uisana (A. J- Wagner) 111 

E. emoda (Pfeifler) 112 

E. cah'doniensis, new species 112 

E. huijcimensis (Poey) 113 

E. hcrmudczi Aguayo & Jaume _„ 114 

E. najazaensis Aguayo & Jaume 115 

E. hriarca (Poey) 116 

E. mdijarina maijarina (Poey) 117 

E. inaijdrina guticrrezi Aguayo & Jaume ... 117 

£. maijarina mirandensis Aguayo & Jaume . 118 

£. hlanesi Clench & Aguayo 118 

Genus Glt/ptcmoda Clench & Aguayo "118 

Ghjptemoda torrei torrei (Henderson) 119 

G. torrei freirei Clench & Aguayo 120 

Emoda (?) dementis Clench & Aguayo .. 120 

References Cited 121 

Index 130 

ABSTRACT 

The present paper is the third in a series deal- 
ing with the Cuban representatives of the land 
prosobranch family Helicinidae. The studies on 
the genus Viana (1968) and Priotrochatella 
(1970) have already appeared. The genus Emoda 
is assumed to have originated in Oriente Province, 
probably from some form of tlie West Indian 
genus Alcadia. From there it spread westward 
across the island into Pinar del Rio Province. It 

Bull. Mus. Comp 



occurs in each of the six provinces, but is absent 
from the Isle of Pines. The specific and sub- 
specific taxa assigned to Emoda were investigated. 
Of the 27 taxa proposed, it is concluded that 16 
are valid. In addition, one new species, Emoda 
caledoniensis, is described and the subgenus 
Ghjptemoda Clench & Aguayo 1950 is raised to 
generic rank. 

INTRODUCTION 

The family Helicinidae appears in two 
widely disjunct areas of distribution cen- 
tered in the Caribbean in the western hemi- 
sphere and in Southeast Asia and the 
Philippine Islands in the eastern hemi- 
sphere. The island of Cuba has an espe- 
cially rich helicinid fauna, possessing no 
fewer than four endemic genera with 
numerous species. In a previous study 
(1968), the present authors have mono- 
graphed the endemic genus Viana, and in 
1970, their monograph on PriotwchateJki 
was published. In this work we take up 
two more such genera, Emoda and Ghjpte- 
moda. In this series of studies, we also plan 
to complete the examination of all Cuban 
helicinids and to investigate the Cuban spe- 
cies of the genera HeJicina sensu stricto, 
EutrochateUa, Alcadia, Lucidella, Cerato- 
discus, and Froserpina. 

Lea (1834b: 161) described the first 
Emoda as Ilelicimi pidcherrima and until 
the appearance of Wagner's work (1907- 
1908), most other new species were in- 
cluded in the genus Ilelicimi. The name 



. ZooL, 141(3): 99-130, February, 1971 



99 



100 Bnlk'tin Miisemn of Comparative Zoology, Vol. 141, No. 3 



Emocla was given by H. and A. Adams in 
1856 to a conglomerate of species which 
had few cohesive affinities. Because of this, 
the name was disregarded by later students 
of Cuban mollusks (Arango, 1878-1880; 
Crosse, 1888), and it first entered into 
general use after Wagner (1907-1908) 
published his extensive monograph. Al- 
though this author used only conchological 
and opercular characteristics, and in spite 
of serious defects in his work, he defined 
the taxon, which he considered a subgenus 
of Alcadia Gray, so successfully that all 
the species he considered to belong to 
E7noda are still considered to do so today. 
He confined the group to Cuba. H. B. 
Baker ( 1922), f)lacing Emocla as a subgenus 
of Scliasicheiki Shuttle\\'orth, examined the 
radulae of several species and chose 
HeUcina silacea Morelet 1849 as the 
type. (See below for a further discussion 
of the generic and subgeneric placement 
of Emocla.) Aguayo & Jaume (1954) in- 
troduced several new taxa and briefly 
reviewed some of the older ones. Twenty- 
seven taxa have been proposed for the 
members of this genus; of these we 
recognize 16 and add one new taxon. Of 
the 17, 11 occur only in Oriente Province, 
3 or possibly 4 in Las Villas, 3 in Cama- 
giiey, and 2 in Pinar del Rio. Havana and 
Matanzas have only the widely ranging 
E. .submarginata, which is also the only 
species occurring in more than one 
province. We have records of its occur- 
rence in each of the six provinces of Cuba. 
All the other forms are found in only a 
single province and most of them in nar- 
rowly limited localities ( Pis. 4, 5; Table 1 ) . 
Emocla is confined to the mainland of 
Cuba and to at least one of its satellite 
cayos or keys (Pi. 4). No species is known 
to occur on the Isle of Pines or elsewhere 
in the West Indies. Emocla probably had 
its origin in the eastern end of Cuba and 
especially in Oriente Province, since many 
more species exist in that area than else- 
where on the island. The most widely 
spread species in Oriente is £. pulcherrima 



and, with no opposing fossil records, it 
might be viewed as an early, if not the 
earliest, form of the genus. E. su])marg,in- 
ata, which, \\'ith little difficulty, can be 
derived from pulcherrima, was in all likeli- 
hood the foriu ^\'hich, because of its ability 
to populate lowland as well as hilly areas, 
spread the genus to other parts of the 
island. Emocla hermudezi and E. najazaen- 
sis, two forms found in Camagijey, the 
province immediately adjacent to Oriente, 
can be easily derived from suhmarginata 
and probably resulted from isolation in the 
mountain ranges of Cubitas and Najaza 
respectively. 

The species foimd in Oriente occupy 
calcareous areas isolated by intei^vening 
volcanic rock. The isolation in the more 
western parts of the island results from 
mountain areas separated by noncalcareous 
lowlands. This isolation must have oc- 
curred during the late Tertiary, probably 
dating from the early Pliocene when Cuba 
more or less attained its present form. Only 
in this way can we account for such sharp 
differentiation as that shown, for example, 
between the two neighboring species ciliata 
and hriarea in southern Las Villas Pro\'ince. 

The only area of clear overlap appears 
to be in tlie eastern part of Pinar del Rio 
Province at the Sierra del Rosaria area of 
the Sierra de los Organos. Here E. sub- 
marginata and sagraiana both occur. The 
small form of typical sagraiana (see below, 
p. 105) is more easily derived from .siib- 
marginata than is the larger form called 
"percrassa." It may be assumed that the 
former then is ancestral to the latter. As 
will be shown, however, isolation of the 
two sagraicma forms did not take place, 
and both, with numerous degrees of inter- 
gradation, occur throughout the range. 

Species of Emoda, unlike Viana, are 
found in each of the six provinces of Cuba 
and occur from the eastern tip near Cabof 
Maisi in Oriente Province to the westem 
limits of the Sierra dc los Organos in Pinar 
del Rio. They are largely confined to tht 
mountain areas, except for one species 



Emoda and Glyptemoda in Cuba • Clencli and Jacobson 101 



with the widest distribution, E. suhmaniin- 
ata (Gray), which also occurs in the low- 
lands. Though Emoda is for the most part 
an upland group, it is not as closely con- 
fined to a limestone substrate as Viana, 
and it is found on the ground under rotting 
leaves, and on branches and vines, as well 
as on calcareous rocks. 

Alcodia striatum (Lamarck) from Puerto 
Rico, the type-species of the subgenus 
Striatemoda H. B. Baker 1940, is obviously 
not an Emoda, despite Baker's doubts on 
this point (1940: 71). Among other dif- 
ferences such as size, operculum, and color, 
it also lacks the diagonal, slightly curved 
axial furrows on the protoconch. The fact 
that it has no spiral sculpture is not signifi- 
cant, since most species of Emoda also lack 
this feature. 

We have not been able to examine or at 
least confirm the location of some pertinent 
type material. Dance (1966) reported that 
the nonmarine shells of Pfeiffer, among 
which were many types of Cuban land 
shells, became part of the Dohni collection 
in the Stettin (Szczecin) Museum, Poland, 
where it was totally destroyed during 
World War II. A personal communication 
(July 1968) from the museum director. 
Dr. W. Filipo\\'iak, confirmed this fact. The 
words "type destroyed" in the text of this 
study reflect this situation. 

The types of the species described by 
Poey and some described by Cundlach. are 
presumably in the Museo Poey^ in the 
University of Havana. In spite of repeated 
requests, we have failed to obtain permis- 
sion from the museum authorities to ex- 
amine the collection, nor has it been 
possible for us to have the presence of this 
material in the collection verified. Never- 
theless, we suppose that the material is 
there and indicate this assumption by 



^ A new public museum has been established 
recently, with exhibits housed in tlie former 
Capitol building in Havana. This museum is 
called the Museo Fehpe Poey and should not be 
confused with the older Museo Poey in the Uni- 
versity of Havana. 



writing "Type, probably MP" in the body 
of the text. 

The specimens examined are in the col- 
lection of the MCZ, unless otherwise noted. 

ABBREVIATIONS USED: 

MCZ Museum of Comparative Zool- 

ogy, Cambridge, Massachusetts. 

USNM United States National Mu- 

seum, Washington, D. C. 

MP Museo Poey, University of Ha- 

vana, Havana. 

BM(NH) British Museum (Natural His- 
tory), London. 

ACKNOWLEDGMENTS 

We gratefully acknowledge the generous 
help of Drs. Harald A. Rehder and Joseph 
P. E. Morrison of the United States 
National Museum, who made much com- 
parative material available to us. Drs. Ken- 
neth J. Boss and Ruth D. Turner, and Mr. 
Richard I. Johnson, all of the Museum of 
Comparative Zoology, read the manuscript 
and offered many valuable suggestions. We 
are also especially grateful to Dr. Turner 
for providing the anatomical and radular 
discussions for the present study. We are 
grateful to Mr. J. F. Peake and Mrs. Angela 
Cane of the British Museum (Natural 
History) for supplying us with the 
BM(NH) catalog numbers of some of 
Orbigny's types and the illustration of the 
type of Helicina crassa Orbigny. Dr. A. 
Riedel of Warsaw, Poland, kindly supplied 
the catalog numbers of two types. Dr. Hor- 
tensia Sarasiia of the Academia de Ciencias 
de la Republica de Cuba most graciously 
provided us ^^'ith excellently presented 
alcoholic material from Pinar del Rio. This 
study is part of the work done under NSF 
Grant No. GB 1004. 

Genus Emoda H. and A. Adams 

Emoda H. and A. Adams 1856, the Genera of 
Recent Mollusca, London, 2: 304 [as a subgenus 
of Helicina Lamarck 1799], (type-species, Heli- 
cina silacca Morelet 1849, subsequent designation 
H. B. Baker 1922: 56). 



102 Bulletin Museum of Comparative Zoology, Vol. 141, No. 3 



Description. Shell ^cncralK about 15 to at times is strongly sculptured. Alcadia is 

30 mm in diameter, imperforate, depressed generally smooth, unicolored, or white with 

to moderately depressed, to subglobose, shades of brown or pink; these various 

dull or moderately glossy and generally colors are never arranged in bands. There 

solid. Color brown, green, yellow, or red- are also constant differences in the oper- 

dish puii-)k\ banded in occasional popu- cula, especially at the raised columellar 

lations or indi\idual specimens. Outer lip edge, which in Alcadia has the upper por- 

simple to considerabK- thickened, generally tion extended internally as a low, narrowly 

witli l)asal notch and /or protuberance, triangular protuberance, a condition not 

Axial sculpture variable: smooth or with found in Emoclo. Finally, Alcadia is pri- 

faint, irregular grow th lines or with regular, marily a Jamaican genus wdth some forms 

irregular, or wrinkU^d ribs. Spiral sculpture in Cuba and elsewhere, whereas Emoda 

generalK- wanting, or when present, con- occurs only on the main island of Cuba, 

sisting of rather faint incised lines or rows The Alcadia closest to Emoda in appear- 

of small pits. Periostracum weak to strong, ance is the species nuda Pfeiffer 1S66, and 

shining or lusterless, wanting in occasional its subspecies, E. n. bagaensis Aguayo 1953, 

species, sometimes laised in somewhat from Oriente. Here, however, an exami- 

hirsute spiral ridges. Operculum auricu- nation of the operculum will immediately 

late, concave, thickest at columellar margin, reveal the true relationship, 

thmning rapidly and becoming bladelike The differences between Schasicheila 

at outer margin. External calcareous layer and Emoda in shell and opercular structure 

of operculum nearly smooth, microscopi- are even more pronounced. The shells of 

cally pitted, white, light browTi, or reddish; Schasicheila are colorless and much more 

internal chitinous layer reddish or brown- fragile, and the roundly inflated base gives 

ish, marked by irregular, semicircular it a shape that is never seen in Emoda. 

growth lines. Columellar edge raised in (See below for further discussion.) It also 

narrow ridge, generally straight but bent differs in the nature of the peristome at its 

inward at top. forming a short, obli(iue insertions in the basal angle and the notch 

furrow intemally. Nucleus lateral and in the upper columellar angle. The wide 

central on colmnellar edge. parietal callus, found in all species of 

Remarks. Emoda has at various times Emoda, is wanting in Schasicheila. Finally, 
been regarded as a subgenus, either of the operculum, though superficially similar, 
Schasicheila Shuttleworth 1852 (Baker differs as follows: in Emoda it always 
1922; Thiel{\ 1929) or of Alcadia Gray presents some color, either in the corneous 
1840 (Wagner, 1907-1908; Keen, 1960). In layer alone or in both calcareous and cor- 
this study, we follow Baker (1926) and neons layers, wdiereas the operculum of 
Wenz (1938) and accord Emoda full Sc/!«.s/c/je//« is always white, with occasion- 
generic rank in the subfamily Helicininae. ally only a small area of light brown near 

Although the shells of Emoda have the columellar ridge. The columellar niar- 

several f(\itures in common with those of gin of the Emoda operculum, as \'iewed 

Alcadia. espccialK' with the subgenus Anal- from the side, is relati\'ely straight or only 

cadia Wagner, they are readily distinguish- very weakly sinuous and is more or less of 

able. In Emoda the shells are generally the same widtli throughout, whereas in 

larger, frequently considerably so, than the Schasicheila it is strongly bent inward at 

Alcadia from Cuba; the smallest Emoda the center, and is noticeably thickest at 

shells are about 15 mm in diameter, the midline and thinnest at both t(>rmi- 

whereas many species of Alcadia have nations. 

shells considerably smaller. Emoda gen- The radulae of E. silacea, sagraiami, and 

erally has a far more vivid coloration and ciliata w^ere described by H. B. Baker 



Emoda and Glyptemoda IX Cuba • Clench and Jacohson 103 



( 1922 ) . In this discussion, we are using 
Baker's concepts and terms as defined on 
his page 30. He found that the radula of 
Emoda is typically that of the subfamily 
Helicininae. It possesses the central tooth 
complex of seven plates (one rhachidian 
and three paired), and a strong capituli- 
form complex consisting of a strong, sub- 
rhomboid comb-lateral plate and a smaller, 
but heavy, accessory plate which is only 
weakh' articulated with the comb-lateral. 
Baker found some variation among the 
three species he studied and felt that ciliafa 
might belong in a special section. The 
paired central teeth and the comb-lateral 
are ^^'ell cusped, \\ith ciliata ha\dng a 
somewhat smaller number of cusps on the 
comb-lateral than the other two species. In 
addition, it has a few more inner bicuspid 
marginals than the other t\\'o: six or se\^en 
instead of four or five. Troschcl (1857: S2, 
pi. 5, fig. 12) described and figured the 
radula of E. suhmarginata in much more 
general outline. His figure shows the ac- 
cessory plate completely separated from 
the comb-lateral, whereas Baker demon- 
strates that the former articulates weakly 
^^■ith the latter. 

Isenkrahe (1867) investigated the anat- 
omy of HeJicina titanica ( = Emoda 
pidcherrima titanica). Bourne (1911), al- 
though he did not specifically study any 
Emoda, was able to make certain recti- 
fications in Isenkrahe's work on the basis 
of his investigations of the closely allied 
genus Alcadia. Baker (1926) provided 
some notes on the anatomy of E. sagraiana 
and E. suhmarginata. Both Bourne (1911: 
777) and Baker (1926: 35) commented on 
the general uniformity of the genitalia in 
the family and the uselessness of these 
characteristics for diagnostic purposes. For 
the present study. Dr. B. D. Turner of the 
Museum of Comparative Zoology dissected 
a specimen of E. sagraiana obtained from 
near Sumidero, Pinar del Bio, which Dr. 
Hortensia Sarasua of Marianao, Havana, 
Cuba, most graciously provided. A com- 
plete report of this investigation ^^'ill be 



published by Dr. Turner later in this series 
of studies. For the moment, we are pro- 
\iding a iew remarks and figures of the 
gross anatomy (PI. 6). 

In alcoholic specimens the color is \&.r\- 
able, ranging from nearly uniformly ivory 
in immature specimens to dark gray or 
black in the adult. The base of the ten- 
tacles and the mantle are dark gray, the 
color gradually growing paler on the dorsal 
part of the foot and becoming almost ivoiy 
at the foot margin and the distal portion 
of the tentacles. The tentacles are long 
and slender, somewhat wider proximally. 
The eye is located on a low peduncle on 
the outer part of the base of the tentacle. 
The sole of the foot is creamy white, 
rounded anteriorly and bluntly pointed 
posteriorly. The columellar muscle at the 
insertion edge is long, curved, cream 
colored, and rounded at the interior end. 
The odontophore is long and cur\^ed and is 
pro\'ided \\i\\\ a short hook at the inner 
termination. Other details of the anatomy 
can be found in Baker (1926: 48). 

Key to the Species of Emoda 

1. Shell large, adult generally 24-32 mm in 

diameter 2 

Shell smaller, adult generally 16-22 mm in 
diameter .10 

2. Shell strongly sculptured 3 

Shell smooth or with weak growth lines 
only 6 

3. Sculpture regular 4 

Sculpture irregular, of strong diagonal 
growth lines, shell yellow SILACEA 

4. Sculpture of prominent spiral furrows, red- 
dish color _. EMODA 

Sculpture of axial lines only 5 

5. Shell green, generally c. 24 mm in diam- 
eter B AYAMEXSIS 

Shell brownish red, generally c. 28-30 mm 
PULCHERRIMA TITANICA 

6. Lip strongly expanded and strongly re- 
flected SAGRAIANA 

Lip weakly expanded, not reflected 7 

7. Shell wine colored, with white band at 

periphery and subsuturally BRIAREA 

Shell without band - — 8 

8. Shell rufous, periostracum oli\aceous 

CALEDONIEXSIS 

Shell yellowish green 9 



104 Bulletin Museum of Comparative Zoology, Vol. 141, No. 3 



9. Shell with spiral lines on periostracum only 

CILIATA CILIATA 

Shell with spiral lines incised on surface .— 
^ CILIATA CUISANA 

10. Distinct spiral as well as axial sculpture 

present - 11 

Spiral sculpture very weak or absent 12 

11. Shell with rounded carina, generally brown- 
ish -- SUBMARGIXATA 

Shell with subacute keel, generalK- faintl\- 
greenish NAJAZAENSIS 

12. Shell with distinct notch at parietal in- 
sertion of upper lip BERMUDEZl 

Shell witliout notch 13 

13. Shell shape turbinate 

PULCHERRIMA PULCHERRIMA 

Shell shape subglobose 14 

14. Sculpture of diagonal, straight, regular 

lines BLANESI 

Sculpture of vermiculate wrinkles 15 

15. Shell small, adult reaching 16 mm 

MAYARINA MAYARINA 

Shell larger, adult reaching 22 mm 16 

16. Sculpture strong, shell reddish green 

MAYARINA GUTIERREZI 

Sculpture narrower, color reddish 

_____ MAYARINA MIRANDENSIS 

Emoda silacea (Morelet) 

Plate 1, figures 20, 21; Plate 5. 

Hdicina silacea Morelet 1849, Testacea Novis- 
sima Insulae Cubanae et Americae Centrahs, 1: 
20 (Cuba; type-locality, here restricted. El Yunque 
de Baracoa, Oriente; 3 svntypes, BM(NH) 1893. 
2.4.813-15). 

Hdicina ochracca Poey 1851, Memorias Historia 
Natural Isla de Cuba, 1: 112, pi. 11, figs. 1-4 
(Baracoa; type, probably MP). 

Ilclicina silacea Morelet, Pfeiffer 1862, Novit. 
Conch., 2: 197, pi. 52, figs. 11-15. 

Alcadia (Emoda) silacea Morelet, Wagner 1908, 
in Martini & Chemnitz, Conch.-Cab., (2) 1: sect. 
18, pt. 2, p. 100, pi. 18, figs. 10-12, 15. 

Description. Shell reaching 28.5 mm in 
diameter, depressed turbinate, rather rough 
and solid. Whorls 4V2, flattened, body 
whorl moderately inflated. Color pale 
greenish yellow, occasional specimens 
showing a faint brownish tinge, the color 
stronger at the interstices than on the axial 
cords. Some specimens with lighter pe- 
ripheral band bordered by subperipheral 
browTi band of varying width; spire and 
base general!}' lighter. Spire moderately 
raised, rounded. Aperture semilunate, outer 



lip unevenly curved, columella white, short, 
weakly convex above, concave below. 
Peristome strongly flaring but \\'eakly re- 
flected above, unevenly thickened, widest 
and thickest at or near the periphery, nar- 
row above where it merges into the parietal 
callus, and below where there is a small 
protuberance of varying shape. Parietal 
callus thin, white in the columellar area, 
transparent near outer margin. Suture 
weakly impressed. Penultimate and body 
\\horls prominently marked by more or less 
irregular, diagonal, somewhat wavy axial 
cords, separated by wider but quite shallow 
intervals. Sculpture weak on the early post- 
nuclear whorls. Protoconch IV2 whorls, 
lighter in color than the rest of the shell, 
flattened, marked with faint, curved, diag- 
onal, axial wrinkles. Periostracum thin and 
lusterless. Operculum as in genus, reddish 
in color, the internal corneous layer darker 
than the external, moderately lustrous 
calcareous lamina. 

Height Diameter 
mm mm 

17.0 28.5 El Yunque, Baracoa, Oriente 
17.0 27.5 El Yunque, Baracoa, Oriente 
15.5 23.5 Mayari, Oriente 
15.5 21.5 Finca "La Caridad," near Baracoa, 
Oriente 

Remarks. This species is well character- 
ised by its large size, prominent diagonal 
axial ornamentation, and striking uniformity 
of color, yellow predominating, with oc- 
casional specimens having a brownish tinge. 
At the type-locality, a large number of 
specimens have a lighter peripheral band 
bordered by a brown subperipheral band 
of varying ^^"idth. The typical forms seem 
to be confined to El Yunque itself, the 
specimens from Mayari being somewhat 
smaller and the ones from Finca "La Cari- 
dad" considerably so. This \'ariation in size 
in different populations is quite common in 
the genus, and may be associated with the 
size of the area inhabited and/or variations 
in other features of the locality, such as the 
availability of food, shelter, etc. E. silacea 
differs from pulclwrriyna tita^iica of about 



Emoda and Glyptemoda in Cuba • Clench and Jacobson 105 



the same size, in color, in being more de- 
pressed, in having a thicker hp and more 
irregular and lower axial costae. 

Poey described ochracea because he 
failed to find the decussate base that More- 
let had mentioned in his description of 
silacea. However, as Pfeiffer pointed out 
(1862: 198), Morelet was in error because 
he did not have a large enough sample 
when he wrote his description and con- 
sidered an occasional ^^ariation to be a 
consistant feature. 

Specimens examined. ORIENTE. El 
Yunque de Baracoa; Mayari; Soledad, 
Guandoa, Baracoa; Finca "La Caridad" on 
road from Baracoa to Duaba. 

Emoda sagraiana (Orbigny) 

Plate 1, figures 16-19; Plate 4, figure 1. 

Helicina sagraiana Orbigny 1842, MoUusques, //! 
Sagra, Histoire Physique, Politique et Natuielle de 
I'lle de Cuba, 1: 240, pi. 18, figs. 12, 13 (I'inte- 
rieur de I'lle de Cuba; Cerro de Cuzco [not "aux 
environs de Trinidad"]; 4 syntypes, BM(NH) 
1854.10.4.163). 

Helicina sagra Sowerby 1847, Thes. Conch., 1: 
3, pi. 1, fig. 10, pi. 3, fig. 126 (Cuba [error for 
H. sagraiana Orbigny]). 

Trochatella (Viana) sagra "d'Orbigny" Chenu 
1859, Manuel de Conchyhologie, Paris, 1: 496, 
fig. 3691 [error for sagraiana]. 

Helicina caialinensis Pfeiffer (July) 1856, Malak. 
Bliit. 3: 56 (prope Catalina, provincia occiden- 
talis insulae Cubae; type destroyed); Pfeiffer 
(Dec.) 1856, Novit. Conch., 1: 83, pi. 23, figs. 
1-6. 

Helicina cataJiniana Sowerby 1866, Thes. 
Conch., 3: 278, pi. 1, fig. 8 [error for caialinensis]. 

Alcadia (Emoda) sagraiana d'Orbigny, Wagner 
1908, in Martini & Chemnitz, Conch. -Cab., (2)1: 
sect. 18, pt. 2, p. 94, pi. 17, figs. "1-4, pi. 18, 
fig. 13. 

Emoda sagraiana percrassa Aguayo & Jaume 
1954, Torreia, 21: 12, pi. 1, figs. 1-3 "(Luis 
Lazo, Provincia de Pinar del Rio; holotype, MP 
17341). 

Description. Shell 17 to 29 mm in diam- 
eter, depressed to moderately raised, gen- 
erally smooth. Whorls about 5^4, veiy 
slightly rounded, body whorl depressed, 
rather narrowly rounded at the periphery. 
Color generally brownish, from pale yel- 
lowish buff to bright orange-brown; oc- 



casional specimens with an olivaceous 
tinge, others with a lighter peripheral band, 
spire and callus frequently reddish. Spire 
moderately raised, rounded, domelike, the 
same color as the rest of the shell, occasion- 
ally reddish, rarely darker, sometimes 
lighter. Aperture widely semilunate, some- 
what extended laterally and more or less 
regularly rounded in peripheral region, 
white near lip, brownish in the interior. 
Lip thickened, occasionally very much so, 
slightly flaring, rarely reflected above. Basal 
callus slightly raised, white in columellar 
region, transparent near outer margin, 
as extensive as, but more rounded than, the 
aperture. Columella short, oblique, white, 
concave below; basal notch usually deep, 
acute or obtuse; basal tooth low, rounded. 
Suture moderately impressed. Axial sculp- 
ture of very fine, irregular, diagonal growth 
lines. Spiral sculpture wanting. Proto- 
conch l^/i \\'horls, sublustrous, slightly 
raised, faintly marked by curved, retractive, 
axial wrinkles. Periostracum thin but strong, 
light or dark brown or olivaceous, in some 
specimens bro\\'n tinged with green. Oper- 
culum as in genus, large, strong, calcareous 
lamina white, or tinged with faint reddish 
bro\\'n, generally on one plane but curved 
sharply and shortly inwardly below; in- 
ternal corneous lamella thin, brownish, 
somewhat darker at the margins. 



La Claraboya, Luis Lazo 

Ensenada de los Biuros, Cabezas 

Caiguanabo, Consolacion del Norte 

Bejarano, Canalete 

Galalon, San Andres 

Catalina, San Diego de los Bairos 

Sierra Paso Real, Guane 

Cafetal "La Villa," Candelaria 

Galalon, San Andres 

Pan de Azucar, Viiiales 

La Muralla, Guane 

Remarks. A highly polymorphic species, 
E. .sagriana varies from a small brownish 
form with an acute basal notch and rela- 
tively thin peristome to large greenish 
forms with a red spire and/or callus and 
an exceedingly heavy lip. In between are 



Height 


Diameter 


mm 


mm 


21.5 


29.0 


17.5 


28.3 


17.5 


26.0 


17.5 


25.5 


16.0 


25.5 


16.0 


24.5 


15.5 


23.5 


15.5 


23.3 


15.0 


23.5 


14.5 


24.0 


13.5 


19.5 



106 BiiUctin Museum of Comparative Zoology, Vol 141, No. 3 



found all kinds of inter2;radcs scattered 
thronghout the range. The smallest forms 
come from around Guane near the western 
terminus of the Sierra de los Organos, but 
almost similar forms are found near Cata- 
lina, toward the eastern end. The hea\'y, 
thicklipped form, to which Aguayo & Jaume 
gave the subspecific name of pcrcrassa, 
occurs generally in the region of Luis Lazo, 
Sumidero, and Cabezas, in the west central 
portion of the Sierra de los Organos; speci- 
mens which cannot be separated from those 
found here also occur, together with the 
smaller, more typical specimens toward the 
east at San Andres, Consolacion del Norte, 
and San Diego de los Baiios. The popu- 
lations at Ensenada de los Burros in Ca- 
bezas and at Finca "La Giiira" near Luis 
Lazo can be readily divided into the small 
brownish form with the acute notch and 
thin lip, and the large, heavy lipped form 
with the obtuse notch and heavy peristome. 
The complete lack of intergrades in these 
localities argues strongly that here are 
two distinct sympatric groups that readily 
deserve specific distinction. These clear 
differences, however, fade away in the 
populations from the eastern end of the 
range around the Sierra San Andres and 
San Diego de los Baiios, where many arbi- 
trary decisions are needed to place the 
majority of the specimens. 

Aguayo & Jaume (1954: 4) well de- 
scribed the confusion that seems to exist: 
"We are confronted by the following 
dilemma: either we name all the \'ariants 
which occupy more or less restricted areas 
( 'microspeeies' or incipient subspecies), 
thus increasing the confusion of the present 
nomenclature, or we put all the diverse 
forms under a single specific name, thereby 
creating the error of postulating a uni- 
formity that is far from the truth" (trans- 
lated). They go on to suggest that only 
the two extreme forms be given names 
(sograiana and percmsso) and that the 
intermediar\^ colonies be designated by the 
"conventional" formula: Emoda sagraiana 
Orbigny tr. percrassa. This is obviously no 



solution, since the determination of just 
where extreme forms end and the inter- 
mediaiy forms begin will have to be largely 
subjective. Moreover, we agree with Mayr 
(1969: 46) who wrote, "Now that it is 
being realized that every local population 
is different from eveiy other one, even if 
they li\'e only a few miles apart or less .... 
there is no longer any excuse for a formal 
recognition of innumerable local subdivi- 
sions of subspecies." On the basis of the 
data and the large amount of material 
available to us at present, we can only 
conclude that we are dealing with a highly 
complex and variable species whose com- 
ponent parts cannot be completely under- 
stood without intensi\e biometric and 
ecological study. For the present, we feel 
that least confusion will result if all the 
Emoda-likc populations in the Sierra de los 
Organos mountain complex are considered 
sagraiana. This interpretation should prove 
as utilitarian as calling all the polymorphic 
populations of the littoral marine species 
Thais lapillus in the Western Atlantic or 
T. JamcUosiis in the Eastern Pacific. 

Specirnens examined. Pinar del Rio. 
Guane: Sierra de Guane; "Sierra Mendoza"; 
Sierra Paso Real; La Muralla; Punta de la 
Sierra. Luis Lazo: Sierra San Carlos; La 
Caoba; Sierra Los Acostas; La Cetricina; 
La Claraboya; La Giiira; El Potrerito; La 
Estrechura; El Junco; Ensenada del Bar- 
rio, Pica Rica, and Mogote La Giiira, 
Sumidero. Cabezas: Valle Isabel Maria; 
Ensenada de los Burros; Sierra La Maje- 
quilla, Petiablanca; Ensenada Chica; Valle 
de Cabezas; Mogote Cementerio; Sierra del 
Quemado. Vinales: Potrero de Manuel 
Sanchez near Capon; Arroyo Melindre; 
Kilometer 14; Mogote de la Vega No. 3; 
Mogote Capon; Sierra Penitencia; Sitio del 
Infienio; Sierra Derrumbada; Sierra Cela- 
das; Sierra Celadas on the road to La Mina 
Matahambre; Mina Matahambre; Hoyo de 
los Cimarrones, Sierra del Infienio; El 
Cuajani; Los Peladores at summit of Sierra 
del Infierno; Sierra del Infierno; Sierra el 
Martillo; El Queque; Ensenada del Valle, 



Emoda and Glyptemoda in Cuba • Clench and Jacobson 107 



El Queqiie; Sierra de Galeras; Mogote 
Pequeilo; Mogote Rojas; Mogote Dos 
Hermanos; Puerta del Ancon; Lorenzo 
Lopez, 1 km N of Vinales; Hoyo Jutia; El 
Cejanal; Hoyo de Fania, Palmarito; Ense- 
nada de Martin Miranda, Palmarito; Hovo 
de Lorenzo Martinez, Palmarito; Las Deli- 
cias; Cueva de los Santos, Palmarito; Sierra 
de Palmarito; Mogote Quillo; El Punto, 
Pan de Azucar; Pan de Azucar; Hoyo 
Largo, San Antonio; Hoyo de los Muertos; 
Mogote Jose Maria Garcia. Smi Vicente: 
La Chorrera; Mogote la Mina Constancia; 
Mogote Jutia, 1 km E of Baiios de San 
Vicente; Bafios de San Vicente; Ensenada 
de los Baiios; Mogote Pequeiio, Costanera 
de San Vicente; Hoyo del Ruisenor, Ancon; 
Cueva del Rio, Baiios de San Vicente; 
Costanera de San Vicente; Mogote de la 
Resbalosa; Mogote del Marmol and Laguna 
de Piedras; Sierra del Abra; Punta de la 
Costanera de San Vicente; Ensenada de 
San Vicente; Las Cuevitas, Ancon; Puerta 
del Ancon. Consolacion del Norte: Monte 
Largo, Pico Chico; Mogote Cascajal near 
Pico Chico; Mogote Grande SW of Pico 
Chico; Puertecitos de Caiguanabo; Los 
Portales, Caiguanabo; Farallon de las Avis- 
pas; Pico Grande; Bejarano, Canalete; Abra 
de Bejarano; "Sierra Canalete"; Sierra San 
Andres; La Sierra; Sitio de la Sierra; 
Mogote de la Cidra; Mogote de la Palma; 
Mogote el Indio; Ensenada de la Ayua; 
Pasada de la Ayua; La Jagua; La Jagua 
de Azquiz; Sierra Gloria, N of Colniillo 
de la Vieja; Colniillo de la Vieja; Los 
Caracoles; Abra de la Colmena; Colmena 
de Piedra, Galalon; Sierra Grande, Gua- 
camaya. San Diego de los Bafios: Mogote 
de la Finca; Mogote el Bosque; La Catalina, 
N of San Diego; Cayito, 3 mi. from La 
Catalina; Hoyada de la Catalina; Mogote 
Colorado; Pinalito; Mogote de los Indios; 
Los Cayitos, Catalina, N of San Diego de 
los Baiios; Cueva del Indio. Guajaihon: 
San Jose de Sagua; SW of Pan de 
Guajaiboii; San Juan de Sagua; Sierra 
Chica; Ranclio Lucas, W of Guajaiboii; 
Sierra de Juan de Guacamaya, Baliia 



Honda; Hato Sagua; Caimito; La Funiia; 
Sierra la Giiira; El Toro, Sierra Limones. 
Cabanas: El Guabinacho; 2 mi. N of Santa 
Cruz; Las Animas, Rangel; El Retiro, 
Rangel; El Maiiiey, Callajabos (Caya- 
jabos); Loma de Gloria; Rancho Mundito, 
Rangel; Subida al Rangel; El Taco. 
Candelaria: Escuela de Frias; Cafetal "La 
Villa." 

Emoda pulcherrima pulcherrima (Lea) 
Plate 1, figures 14, 15; Plate 2, figures 13, 
14; Plate 3, figures 6, 7; Plate 5. 

Helicina pulcherrima Lea 1834, Trans. Ameri- 
can Phil. Soc. (NS), 5: 49, pi. 19, fig. 57 (Java?); 
1834, Observations on the Genus Unio, Phila- 
delphia, 1: 161, pi. 19, fig. 57 ("Java?"; lecto- 
type, probably figured specimen, here selected, 
USNM 104613; paralectotypes, USNM 104613a, 
104609, 104595, all from Lea collection; type- 
locahty, here selected, Guantanamo). 

Helicina crassa Orbigny 1842, Molliisques, in 
Sagra, Histoire Physique, Politique et Naturelle de 
rile de Cuba, 1:'243, pi. 19, fig. 5 [not fig. 6] 
(interieur de I'lle de Cuba; type, BM(NH) 
1854.10.4.167). 

Helicina pulcherrima Lea, Pfeiffer 1850, in 
Martini & Chemnitz, Conch. -Cab., (2) 1: sect. 
18,. pt. 1, p. 17, pi. 1, figs. 5-7, pi. 6, figs. 5, 6 
(die Insel Cuba). 

Helicina ruhrocincta Poey 1854, Memorias His- 
toria Natural Isla de Cuba, 1: 417, pi. 33, figs. 
16-19. ( "Probablemente en el departmento ori- 
ental"; type, probably MP.) 

Description. Shell about 18 mm in diam- 
eter, depressed globose, strong. Whorls 4y2, 
flat, body whorl descending shortly near 
the aperture. Color varied, light green with 
darker green, irregular, faintly sigmoid, 
axial bands, or light to dark reddish brown; 
spire usually lighter. Occasional specimens 
have a narrow, light-colored peripheral 
band. Spire moderately raised, rounded. 
Aperture widely semilunate, white inside; 
peristome thickened, moderately reflected, 
widest at center, gradually narrowing at 
extremities, with a small, elongate, rounded, 
toothlike protuberance near the basal 
notch. Columella short, white, barely con- 
vex, more strongly so below the mid-point. 
Parietal callus strong, as wide as the aper- 
ture, white in the columellar region, darker 



108 Bulletin Museum of Comparative Zoology, Vol. 141, No. 3 



Height 


Diameter 


mill 


mm 


15.1 


20.6 


15.5 


21.5 


13.5 


19 


13.0 


19 


11 


16.5 



near thc> outer, rounded edge; in oecasional 
speeimens the entire callus is white, a pos- 
sible gerontic sign. Body whorl sculptured 
with strong, regular, rounded axial cords, 
wider than the inter\'als, weaker and less 
n^gular on the base and the spire. Proto- 
conch IV2 whorls, moderately raised, with 
irregular, curved axial wrinkles. Periostra- 
cum strong, shining, usually wanting on 
earh' whorls. Operculum with white or 
light reddisli calcareous lamina, and brown 
or rcxldish corneous one. 



Lectotype ( USNM ) 
Mayari, Oriente 
Monte Turquino, Oriente 
Los Arroyos, Miranda, Oriente 
Loma del Gato, Guantanamo, 
Oriente 



Remarks. This subspecies is confined to 
the entire southern portion of Oriente 
Province. It is characterized by its gener- 
ally small size, green or brown color, and 
especially by the strong, regular axial 
sculpture on the body whorl. Even brown 
specimens, which superficially resemble 
suhmar<iinata, can be distinguished by 
their sculpture. Near the Loma del Gato 
in Guantanamo, there is a well-marked 
race of small specimens in which the color 
is grayish green or pale olivaceous with a 
light-colored spire and strong basal callus. 
They have the characteristic strong axial 
ridges. 

We agree with Pfeiffer (1850: 17) that 
cro.^sa Orbigny is synonymous with pul- 
cherrima. Orbigny wrote, "fortement striee 
dans le sens de I'accroissement" (1842: 
243), which is characteristic of pulcherrimo 
(pi. 3, figs. 6, 7). We have no way of 
judging crassa from Cajo [sic] Seco, "west- 
lich von Guba" (Wagner, 1907: 98, pi. 19, 
figs. 5, 6). This is certainly an error in 
locality. 

Poey gave the name ruhrocincto to speci- 
mens of pulcherrimu with a rather vivid 
red and white peripheral band. But in pul- 
cherrima, as in other species of Emoda, 
banded and unhanded forms exist in the 



same populations. Lea (1834b: 162) noted 
that the species had an "indistinct white 
and brown band," and in his remarks, he 
mentioned the fact that the band some- 
times consists of a white line only. We 
therefore follow Wagner (1907: 97) in 
recognizing ru])rocinctu Poey as a synomym 
of piilcherrima. 

The lectotype, here selected, appears to 
be the specimen figured by Lea. Both have 
the same dimensions and the descriptions 
agree. The copied label in the USNM read 
"Guba," though Lea's (erroneous) locality 
was "J^va." No original Lea label was 
fovmd. 

Specimens examined. Oriente. Niquero; 
Road to Mora; La Vigia, E of Mora; Pico 
Turquino; Loma del Gato; Mountains N 
of Imias; Finca "Santa Teresa," Barrio 
Florida, Songo; Los Arroyos, Miranda; 
Banabacoa; Dos Bocas, N of Santiago de 
Guba (USNM); Santa Maria de Loreta, 
near Tiarriba (USNM); Loma Gran Piedra, 
N of Santiago de Guba (USNM); between 
Aguadores and Las Lagunas, E of Santiago 
de Guba (USNM); loma (hill) E of 
Siboney (USNM); road from Loma de la 
Bandera near Pinar de Mayari; woods at 
Arroyo Blanco, Mayari (USNM); Ganapa, 
Mayari; El Purio, Mayari. 

Emoda pulcherrima titanica (Poey) 
Plate 2, figures 8, 9; Plate 5. 

Hclicina titanica Poey 1851, Memorias Historia 
Natural Isla de Cuba, 1: 110, pi. 11, figs. 13-16 
(las costas de Baracoa [Oriente]; type^, proliably 
MP). 

Hclicina titanica Poev, Pfeiffer 1862, Novit. 
Conch., 2: 196, pi. '52, figs. 6-10 (Mala 
[Baracoa] ). 

Alcadia {Emoda) pnlchcrrinia titanica Poey, 
Wagner 1908, in Martini & Chemnitz, Conch. - 
Cab., (2) 1: .sect. 18, pt. 2, p. 99, pi. 17, figs. 
13-15, pi. 18, fig. 16. 

Description. Shell like that of pidcher- 
rima but differing from the nominate form 
in its generally considerably larger size, 

^ The type-specimens can be identified by the 
crab hole ( 10 mm X 5 mm ) in the columellar 
region (Poey, 1851: 111). 



Emoda and Glyptemoda in Cuba • Clench and Jacobson 109 



which leaches 29 mm in diameter. The 
color is more uniformly brownish; oc- 
casional specimens have an olivaceous 
tinge. The characteristic strong axial sculp- 
ture is easily apparent, even to the naked 
eye. 



Height 


Diameter 




mm 


mm 




23.0 


29.0 


Baracoa, Oriente 


21.0 


28.0 


Mandinga, Baracoa, Oriente 


20.5 


26.0 


Barigua, Oriente 


18.0 


25.0 


Baracoa, Oriente 


16.5 


21.0 


Veguitas, Baracoa, Oriente 



Remarks. This subspecies is limited to 
the area around Baracoa, Oriente. This 
distribution and its considerably larger size 
would be enough to regard it as a full 
species, but the only difference in shell 
morphology from pulcherrima pidcherrima 
is its size. Consequently we concur with 
\^^agner in considering it a subspecies. One 
lot of four specimens from Veguitas, Bara- 
coa, can almost be regarded as intermediate 
in size between p. pidcherrima and p. ti- 
tanica. 

Poey described this form from a series 
of dead shells, each of which had a hole in 
the columellar region. Poey thought the 
shells had been perforated by the animal 
to provide more space for its oversized 
body (1851: 110). Pfeiffer (1S62: 197) 
correctly identified the holes as having 
been made by hermit crabs, which inhabit 
empty shells. (See also Arango, 1878: 87). 

Specimens examined. Oriente. Baracoa: 
Veguitas; Mandinga; Barigua; Boca de 
Yumuri; La Cantera de Miranda; Mata; 
Manglito, Barigua. 

Emoda submarginata (Gray) 

Plate 1, figures 1, 2; Plate 4, figure 2; Plate 
5. 

Hclicina siihmarginaia Gray 1825, Zool. Jour., 
1: 68, pi. 6, fig. 11 (no locality given; type- 
locality, here restricted, Sierra El Grillo, near 
Cotorro, Havana Province, Cuba; tvpe,' BM(NH) 
?). 



^Peake writes (in litt.) that the BM(NH) has 
several lots of specimens, but nothing to identify 
any as the type. 



Heliciiia rubra Pfeiffer 1839, Wiegmann, Arch. 
Naturg., 5th year, 1: 355 (Cuba); type destroyed. 

Helicina sithinarglnata Gray, Orbigny 1842, 
Mollusques, in Sagra, Histoire Physique, Politique 
et Naturelle de I'lle de Cuba, 1: 244, pi. 19, fig. 
8 [not 7] (I'interieur de I'lle de Cuba). 

Helicina suhmarpinata Gray, Pfeiffer 1850, in 
Martini & Chemnitz, Conch.-Cab., (2)1: sect. 18, 
pt. 1, p. 16, pi. 1, figs. 8-10 (Matanzas, Retiro). 

Alcadia (Emoda) pulcherrima planospira Wag- 
ner 1908, in Martini 6< Chemnitz, Conch.-Cab., 
(2) 1: sect. 18, pt. 2, p. 98, pi. 19, figs. 8-10 
(die Insel Cuba; type in Warsaw Zoological 
Museum 8338). 

Emoda zaijasi Aguayo & Jaume 1954, Torreia, 
No. 21, p. 12, pi. 2, fig. 7 (Sierra del Crista], 
Sagua de Tafiamo, Provincia de Oriente, Cuba; 
holotype, MP 17366). 

Description. Shell 15 mm to 20.5 mm in 
diameter, depressed conic, strong, barely 
carinate. Whorls 5V2, flattened, body whorl 
barely descending at aperture. Color varied, 
yellowish olivaceous with light green spire, 
or reddish with a pale orange base, or 
reddish brown with paler spire and base, 
usually with light colored, narrow periph- 
eral band. Spire moderately raised, broadly 
depressed conic or low turbinate, generally 
same color as rest of shell. Aperture sub- 
lunate, palatal lip thickened, barely ex- 
panded, with a shallow notch just in front 
of a low basal tooth. Columella white, 
gently concave, barely rounded at insertion 
in basal portion of lip and extended into a 
triangular lamella that ends at the basal 
tooth. Parietal callus porcellanous, as wide 
as the aperture, ground color of shell 
showing through on outer margin. Sculp- 
ture of variously strong, diagonal growth 
lines and a few widely spaced, concentric 
spiral lines, seen as slightly raised cords in 
the periostracum. These periostracal cords 
are reflected as lines of minute pits in the 
shell surface. Protoconch IVa whorls, usu- 
ally lighter in color than the rest of the 
shell, faintly marked with gently cui-ved, 
diagonal, irregular axial wrinkles. Perios- 
tracum thin, usually absent on upper 
whorls. Operculum as in genus, calcareous 
layer lustrous, thin, white, internal corneous 
lamina orange-brown. 



110 Bulletin Museum of Comparative Zoology, Vol. 141, No. 3 



Height Diaiiifter 
mm mm 

13.0 20.0 Yauniajay, Las Villas 

12.5 20.5 El Crillo, Havana 

12.0 18.0 Piedras de Ainolar, Escanibray, 

Las \'illas 
11.5 15.0 El Rc'tiro, Matanzas 

Remarks. This spccic^s ranges from the 
eastern' part of Pinar del Rio Provinee at 
Artemisa to Mayari in Oriente and thus 
has the most extensive range of any Emoda. 
The distribution is not eontinuous. This 
discontinuitN' is probabK' due to ineomplete 
eollecting in the intervening areas. We 
have but a single reeord of this species 
from Camagiiey. This province, however, 
is probably the least collected area in Cuba. 
In Camagiiey, submariiinata is replaced in 
two important localities. Sierra Najaza and 
Sierra Cubitas, by the closely related spe- 
cies najazaensis and hermudezi, respec- 
tively. 

E. sid)marginota is quite uniform through- 
out its range. The color varies somewhat 
from dark brown to light brown, and some 
olivaceous specimens appear. Occasional 
specimens have a light colored peripheral 
band. The incised spiral cords vary in 
strength but some indications of them are 
usualK' present. There is likewise little 
significant variation in size. The smallest 
specimens are from El Retiro, Matanzas 
(15 mm in diameter), and the largest from 
Havana (20.5 mm), but the great majority 
vary between 18 and 20 mm. 

The color, weak axial ribs, and the pres- 
ence of spiral lines on pidcherrima plano- 
spira Wagner place this form in the 
synonymy of suhmarginata rather than in 
pidcherrima. 

Specimens examined. Pinar del Rio. 
Las Canas; Las Mangas, Artemisa; foothills 
of Rangel. Havana. San Antonio de los 
iJanos; Sierra El Grillo ( t>^pe-locality ) ; 
Sitio Pcrdido; Loma Camoa; Mantilla; 



^Forcart (1950: 71) identified E. suhmarginata 
from Pinar del Rio between the city of Pinar del 
Rio and Viiiales. We have not seen the specimens 
lint suspect that the identification is in error. 



Madruga; El Ingles; La Canabrava; Finca 
"La Belianza," Cotorro; Pefia Blanca, Sierra 
Anafe; Jaruco. Matanzas. Elena, El 
Palenque; Cotorra; Dos Cecilias, Coliseo; 
Jaguey Grande; Abra de Figueroa, Valle de 
Yumuri; Cueva el Negro, San Miguel; 
Canasi Rd., near El Pacito; Loma del Fine, 
Cardenas; Bridge over R. Canimar, Matan- 
zas-Limonar Rd.; 20 mi. SE of Matanzas; 
Ceiba Mocha; Vista Alegra (USNM). Las 
Villas. Soledad: Mogote el Palmar; Bo- 
tanical Garden; Guabairo; Loma de la 
Jutia, Vilche's Potrero; Sagua la Grande; 
Mina Carlota, about 12 mi. E of Soledad; 
Laza del Medio; El Mamey; San Lorenzo; 
Jumagua; Mogote W of La Jumagua. 
Trinidad: Northwest of La Vigia; Sierra de 
Santa Rosa. Yaguajaij: Yaguajay-Remedios 
Rd.; Cerros de Yaguajay; Guainabo; El 
Tigre; Punta Judas; Sierra Judas de la 
Cumagua. Remedios: El Mamey; Los Hatil- 
los; La Culebra; Loma Caraballo; Dos 
Sierras; Buenavista; La Puntilla (USNM); 
Cerro Guajabana. Caiharien: Finca "Cueta"; 
Loma de Petrera; Punta Caguanes, Buena- 
vista Bay; Santa Rosa; Cayo Conuco; Cayo 
Lucas; San Joaquin, Esperanza; El Boque- 
ron, Jatibonico; San Agustin, Zulueta; 
Piedras de Amolar, Escanibray; Villa Ciba, 
Carretera de Camajuani; Chicharron, Sierra 
Vega Alta; La Vigia, Mayajigua; Sierra de 
Canoa, Los Llanados; Monte Cagiieiras, 
Sancti Spiritus; Sierra 3 km S of Dolores. 
Camaguey. Punta Alegre. Oriente. Camino 
de la Loma de la Bandera, Pinar de Mavari; 
Mayari; between Sagua de Tafiamo and 
Cananova. 

Emoda ciliato (Poey) 

Plate 1, figures 7-9; Plate 4; figure L 

Hcliciua ciliaia Poey 1851, Memorias Historia 
Natural Isla de Cuba, 1: 109, pi. 11, figs. 5-8 
(Trinidad [Las Villas]; type, probalily MP). 

Helicina fossulata Poey 1857, Memorias Historia 
Natural Isla de Cuba, 2: 25 (Trinidad, [Las 
Villas]; syntypes, probably MP; paratype, MCZ 
263901 from the Andiony Collection collected by 
Dr. Gutierrez, San Juan de Letran, Santa Clara 
[ = Las Villas]). 

Alcadia {Emoda) ciliata Poey, Wagner, 1908, in 



Emoda and Glyptemoda in Cuba • Clench and Jocobson 111 



Martini & Chemnitz, Conch.-Cab., (2) 1: sect. 
18, pt. 2, p. 102, pi. 8, figs. 5-6; pi. 19, figs. 
14-17 (Trinidad und Banao auf Cuba). 

Description. Shell reaching 27 mm in 
diameter, strong, widely turbinate, smooth 
but lusterless, rounded at periphery. Whorls 
5V2, moderately convex, body whorl 
rounded at base, barely descending near 
aperture. Color predominantly light yel- 
low, occasional specimens faintly tinged 
with brown; white when decorticated, with 
an indistinct, whitish subsutural band con- 
tinued subperipherally on the body whorl. 
Spire depressed, dome shaped. Aperture 
widely semilunate, yellowish within. Palatal 
lip white, strongly thickened, moderately 
flaring. Basal denticle small and obtuse. 
Parietal callus smooth, thick, white in the 
columellar region, transparent near the 
rounded outer margin, more rounded than 
the aperture. Columella short, oblique, 
very weakly sigmoid, angled above, slightly 
rounded at insertion with basal lip. Su- 
ture moderately impressed. Shell smooth, 
marked by very fine, curved, diagonal 
growth lines. Protoconch VA whorls, barely 
raised, moderately lustrous, approximately 
same color as rest of shell, marked by fine, 
curved, axial wrinkles. Periostracum strong, 
closely marked by irregular spiral cords of 
raised periostracal material, not reflected in 
the shell surface beneath. Operculum as in 
genus, light to dark reddish brown near the 
outer margin. 



Height Diameter 
mm mm 



19.0 27.0 Banao, Las Villas 
19.0 25.5 Ciegos de Ponciano, Las Villas 
15.5 19.5 San Juan de Letnin, Las Villas; 
paraty^e of jossulata. 

Remarks. This relatively large-shelled 
species is readily distinguishable from the 
others of more or less equal size by its 
uniformly yellow color and depressed 
shape. It differs from silacea, the other spe- 
cies in which yellow predominates, by its 
smooth, almost unsculptured surface. It 
apparently has a limited range in the moun- 
tains between Trinidad and Sancti Spiritus 



in the southeastern part of Las Villas 
Province. 

Helicina fossidata Poey is only a color 
form with a peripheral band. Poey wrote 
of ciliota, "Color es ya de un amarillo bajo, 
ya de un moreno rojizo," and of jossulata, 
" . . . . sordide rubra . . . cingulis duobus 
flavidis ornata, quorum alter ad suturam, 
alter ad peripheriam." The type locality 
of both forms is "Trinidad" and the only 
difference seems to be the presence of a 
spiral peripheral band in jossulata. Else- 
where we show that banded and unhanded 
forms frequently occur in single popu- 
lations. 

Specimens examined. Las Villas. Banao, 
Sancti Spiritus; Ciegos de Ponciano; 
Camino de Banao, Sancti Spiritus; San 
Juan de Letran. 

Emoda ciliata guisana (A. J. Wagner) 

Plate 1, figures 10, 11. 

Alcadia (Emoda) ciliata p.uisana A. J. Wagner 
1908, in Martini & Chemnitz, Conch.-Cab., (2) 
1: sect. 18, pt. 2, p. 102, pi. 18, figs. 7-9 (Guisa 
auf Cuba; types in Warsaw Zoological Museum 

8353). 

Description. Shell like that of ciliata, 
differing from the nominate form by the 
presence of shallowly incised, parallel, 
spiral lines on the shell, as in suhmar<i,inata. 
These lines seem to correspond to perios- 
tracal structures found in ciliata ciliata, 
but in that subspecies they do not affect 
the shell, being found in the periostracum 
only. Color yellowish, as in ciliata, but in 
some specimens there is a faint reddish 
tinge. 

Height Diameter 
mm mm 

18.5 25.5 "Cuba" 

18.0 25.0 [Guisa, Bayamo, Oriente] (?) 

Remarks. Wagner stated that this sub- 
species came from "Guisa auf Cuba." Some 
students have taken this to mean Guisa, 
near Bayamo in Oriente Province. This is 
most unlikely, since ciliata ciliata, whose 
distribution is limited to central Las Villas, 
has not been found in Camagiiey Province, 



112 Bulletin Muscuiti of Conipaiaticc Zoology, Vol. 141, No. 3 



which separates Las Villas from Oriente 
Pro\ince. Either there is a Guisa — a small 
finca or miniseule s(>ttlcMnent — in Las Villas, 
or Wagner's locality is in error. The mili- 
tar\- map of Cuba cites no Guisa from Las 
Villas. We are keeping this name pro\'ision- 
ally since the material at hand is not suffi- 
cient to indicate whether the incised spiral 
lines are characteristic of individuals or of 
an entire colony. 

Specimens examined. "Cuba"; Guisa. 

Emoda emoda (Pfeiffer) 
Plate 2, figures 6, 7; Plate 5. 

Hclicina emoda Pfeiffer 1865, Novit. Conch., 2: 
253, pi. 64, fijis. 6-8 (parte oriental! ins. Ciibae; 
[Monte Toro, based upon Pfeiffer 1860: 77]; type 
destroyed ) . 

AJcadia (Emoda) sagraiana emoda Pfeiffer, 
Wagner 1908, in Martini & Chemnitz, Conch. - 
Cab., (2) 1: sect. 18, pt. 2, p. 96, pi. 17, figs. 
7-9. 

Description. Shell reaching 31 mm in 
diameter, moderately strong and relatively 
smooth. Whorls 5V2, moderately rounded, 
body xv'horl depressed, descending shaiply 
at aperture. Color of body whorl bright 
green, reddish brown when decorticated; 
spire reddish, peripheiy banded by a 
\\'hitish line margined above and below by 
dull, narrow, reddish bands. Spire rela- 
tively high and dome-shaped. Aperture 
wideK' semilunate, white inside; peristome 
thickened, moderately flaring, not reflected, 
basal tooth generally weak. Parietal callus 
smooth, rounded, larger than the aperture, 
white near the columella and transparent 
near the outer margin. Columella short, 
white, slightly sigmoid, the wider cun'e 
below, and briefly rounded near the basal 
tooth. Suture moderately impressed, rough- 
ened by the terminations of the growth 
lines. Sculpture of low, irregular, mod- 
erately strong growth lines crossed by im- 
pressed, closely spaced spiral furrows, most 
closely approximated at the base and 
weaker on upper whorls. Protoconch VA 
whorls, reddish, smooth, faintly marked by 
irregular, curved, diagonal axial riblets. 



Periostracum strong, green on body vvhorl, 
weaker and reddish on spire. Operculum 
as in the genus. 

Height Diameter 
mm mm 

23.5 31.0 Manatial, Santa Fe, Oriente 
22.0 30.5 Cuaso, Guantanamo, Oriente 
18.0 28.0 El Codrito, Monte Libano, Oriente 

Remarks. This is a well-marked species 
from near Guantanamo in Oriente Province. 
Pfeiffer thought it might be related to 
hriarea from Las Villas Province because 
of its large size, but emoda has a more 
elevated shape, the shell is not so solid, the 
color differs markedly and the spiral sculp- 
ture is completely wanting in hriarea. 
Wagner (1907: 96) confused it with the 
larger sap'aiana from Luis Lazo in Pinar 
del Rio Province, but that shell is some- 
what smaller, more depressed, heavier, has 
a far thicker, more reflected lip, and lacks 
the spiral sculpture. Pfeiffer's figures are 
of decorticated, weathered specimens but 
they show (especially his fig. S) the spiral 
sculpture quite clearly. This species is 
remarkably constant in its color but, like 
many Emoda, it varies in size. The popu- 
lations from Monte Libano consist of 
smaller shells which, however, clearly be- 
long to emoda because of the elevated 
shape, the green and reddish color, and the 
presence of spiral lines. Although the type- 
specimens have been destroyed, there is no 
doubt regarding the identity of this species. 
Pfeiffer's figures are unusually good and 
we have examined several lots from near 
the type-locality. 

Specimens examined. Oriente. La Subida 
de la Hembrita; Guantanamo; El Codrito; 
Monte Libano; Montaiia de Guantanamo; 
Manatial de Santa Fe; 1 mi. NW of Guaso; 
N of Guantanamo City. 

Emoda caledoniensis, new species 
Plate 2, figures 11, 12; Plate 5. 

Ilolotype, MCZ 261352, from Mina Cale- 
donia, Mayari, Oriente. L. R. Rivas, col- 
lector. 



Emoda and Glyptemoda in Cuba • Clench and Jacobson 113 



Height 


Diameter 


mm 


mm 


21.5 


30.5 


21.0 


28.0 


20.0 


29.0 


19.5 


2S.0 


17.0 


24.0 



Para types, MCZ 261350; USNM 463790, 
both from the same locaHty as the holotype; 
MCZ 261348, from Brazo del Pino, Wood- 
fred, Mayan', Oriente; MCZ 261351 from 
Sierra del Cristal, Mayan', Oriente. 



Brazo del Pino, Paratype 
Brazo del Pino, Paratype 
Sierra del Cristal, Paratype 
Mina Caledonia, Holotype 
Mayari, 1 km from the pine forest, 
Paratype 

Description. Shell reaching 30.5 mm in 
diameter, inflated, smooth and moderately 
solid. Whorls 5V2, moderately rounded, 
body whorl inflated, weakly shouldered 
obliquely below the suture, rounded at the 
periphery, shortly descending near the 
aperture, where it blends into the basal 
callus. Color olivaceous, with irregular 
darker axial streaks, light brick red in 
decorticated specimens. Spire moderately 
raised, rounded, darker than rest of shell. 
Aperture semilunate, lip moderately thick- 
ened, flaring, barely reflected above, widest 
near the center, narrowing at each ex- 
tremity; basal tooth small, rounded, basal 
notch barely perceptible. Columella short, 
white, slightly sigmoid and rounded below. 
Basal callus smooth, larger than the 
aperture, white in columellar region, trans- 
parent at outer margin. Axial sculpture of 
very fine, diagonal growth lines, spiral 
sculpture wanting. Suture moderately im- 
pressed, somewhat roughened by the ter- 
minations of the growth lines. Protoconch 
IV2 whorls, faintly marked by cui-ved, axial 
wrinkles. Periostracum strong, olivaceous 
or green. Operculum as in genus, calcare- 
ous layer white, inner corneous layer 
reddish, darker at margins. 

Remarks. The four lots in MCZ upon 
which this description is based all come 
from the region around Mayari in Oriente 
Province. This new species belongs to the 
group of large, predominantly greenish or 
olivaceous shells with a reddish spire from 
the eastern part of the country. It differs 



from pulcherrima titanica from Baracoa by 
its higher spire and in the absence of the 
strong, regular, axial sculpture; it is gener- 
ally smaller and more rounded than E. 
emoda from Guantanamo and lacks the 
spiral sculpture of closely set incised lines. 

Emoda bayamensis (Poey) 
Plate 2, figures 4, 5; Plate 5. 

Helicina bayamensis Poey 1854, Memorias His- 
toria Natural Isla de Cuba, 1: 416, pi. 33, figs. 
7-10 [not 8-11] (Bayamo [Oriente]; lectotype, 
here selected, MCZ 73855, Buena Vista, Bayamo, 
Oriente, Anthony Collection, from Poey; para- 
lectotype MCZ 256496, same locality. The shells 
in MCZ from which the type selections were made 
were identified as type material by Torre in June 
1912). 

Helicitm hastklana Poey 1854, Memorias His- 
toria Natural Isla de Cuba, 1: 415, pi. 33, figs. 
11, 12 (Puerto Principe [ = Camagiiey]; type, 
probably MP). 

Helicina bayamensis Poey, Pfeiffer 1862, Novit. 
Conch., 2: 199, pi. 53, figs. 1-5. 

Helicina jeannereti Pfeiffer 1862, Malak. Bliit., 
9: 6 (Mayari [Oriente]; type destroyed). 

Helicina bayhamensis "Poey" Reeve 1873, 
Conch. Icon., Helicina, 19r pi. 2, fig. 11 (Cuba) 
[error for bayamensis]. 

Helicina bastidiana "Poey" Reeve 1873, Conch. 
Icon., Helicina, 19: pi. 8, fig. 61 (Cuba) [error 
for bastidana]. 

Description. Shell reaching 23 mm in 
diameter, depressed globose and solid. 
Whorls 4^4, weakly convex, body whorl 
moderately inflated, depressed but not 
carinate at the peripheiy. Color of body 
whorl olivaceous and sometimes tinged 
with brown, with narrow, irregular, dark 
green axial stripes; spire much lighter; 
occasional specimens with a yellowish 
peripheral band generally margined with 
red. Spire moderately raised, rounded, 
lighter in color than the rest of the shell. 
Aperture semilunate, irregularly rounded, 
inside margin of upper lip diagonal, almost 
straight. Lip white, well expanded, but not 
reflected, unevenly rounded, somewhat ex- 
tended just above the periphery, widest 
centrally, narrowing rapidly at the extremi- 
ties; basal tubercle small. Columella short, 
white and weakly sigmoid. Parietal callus 



114 Bulletin Museum of Comparative Zoology, Vol. 141, No. 3 



smooth, thin, \\'hite in the coluniellar 
region, transpaic>nt near outer margin. 
Suture moderately impressed. Sculpture of 
irregular, curved, axial cords, \\'eaker and 
less regular than in pulcherrima, weaker on 
the base, almost obsolete on earlier ^^'horls. 
Operculum and periostracum as in the 
genus. 

Height Diameter 
mm mm 

15.0 2.3.0 BiKMia Vista, Bayamo, Oriente, 

Lcctotype 
14.0 21.5 Paralectotype 

Remarks. Poey described hastidana and 
hayamensis at the same time, but in 1S65 
(p. 144) he selected lianamensis as "el tipo 
comun" and wrote that hastidana "es una 
\^ariedad, y muy rara." It seems to be noth- 
ing but a color variant. 

This species differs from the larger 
Emoda by the irregular rather than 
rounded inner edge of the periostome. 
There is a distinct rounded angle above 
and below the periphciy; in other Emoda 
this entire area is evenly rounded. In fully 
adult shells there is also an extension of 
the central portion of the outer lip, so that 
the shell has a somewhat lengthened rather 
than a rounded appearance. This feature is 
illustrated by Poey (1854, pi. 33, figs. 11, 
12) and is present in the lcctotype. Sub- 
mature shells generally lack this character- 
istic. The species differs from pulcherrima 
titanica in its smaller size and weaker axial 
costae. It differs from ciliata in its lively 
colors and comparatively stronger axial 
sculpture, and from sagraiami from the 
other end of the island, in its far weaker 
basal notch and tooth. We could discover 
no feature to distinguish it from jeannereti. 
Pfeiffer, in his discussion of the latter, 
noted several differences between jean- 
nereti and sagraiana but failed to compare 
his species with hayamensis. Wagner 
(1907: 99) cited this as a subspecies of 
pulclierrima, but it lacks the strong axial 
sculpture on the body whorl that is char- 
acteristic of pulcherrima and, at the present 



state of our knowledge, is probably best 
regarded as a species. 

We have not succeeded in clearly identi- 
fying jeannereti Pfeiffer. The species has 
never been figured satisfactorily. Reeve 
(1874: pi. 1, fig. 3a) had a poor figure of 
a shell with a bright green color which he 
referred doubtfully to jeannereti. This 
figure seems to have been copied from 
Sowerby (1847: pi. 3, fig. 126), where it 
appears as Helicina sagra [sic]. Sowerby 
(1866, pi. 1 [266], fig. 5) figured a typical 
brown sagraiana and referred jeannereti 
doubtfully to it. The green form (fig. 6) 
he now called hastidiana (sic) [ = haya- 
mensis]. Arango (1879: 46) copied Pfeif- 
fer's data but cited Wright as the collector 
instead of Jeanneret. Wagner ( 1907-1908 ) 
omitted the name from his monograph and 
Fulton (1915) failed to note this omission. 
The only specimens available for our in- 
spection come from Palma Soriano (Ber- 
mudez collection), about 60 km southwest 
of Mayari, Pfeiffer's type-locality of the 
species. We do not know who made the 
determination. In his discussion, as we 
have noted, Pfeiffer did not compare his 
species with hayamensis. Hence our plac- 
ing jeannereti in the synonymy of haya- 
mensis is provisional. 

Specimens examined. Oriente. La Can- 
tera de Miranda; Buena Vista, Bayamo 
(type-locality); Guisa, Bayamo; Miranda, 
Palma Soriano. 

Emoda bermudezi Aguayo & Jaume 

Plate 1, figures 5, 6; Plate 4, figure 1. 

Emoda hcrmudczi Agiia\o & Jiiinne 1954, Tor- 
reia. No. 21, p. 10, pi. l, figs! 4, 6 ("Paso de 
las Trincheras," Sierra de Culiitas, Provincia de 
Camagiiey; holotype, MP 17346). 

Description. Shell reaching 19.5 mm in 
diameter, rather solid, depressed conic, 
periphery moderately carinate. Whorls AVz, 
flat, body whorl descending rapidly near 
aperture, where it blends into the thick- 
ened, slightly raised basal callus. Color 
generally olivaceous at base, tinged with 



Emoda and Glyptemoda in Cuba • Clencli and Jacobson 115 



brownish orange above, occasional speci- 
mens with a narrow, yellowish peripheral 
band. Spire low, rounded, same color as 
rest of shell. Aperture widely semilunate; 
peristome white, thickened, very weakly 
expanded. There is a distinct notch above 
near the parietal margin from which a 
short, raised marginal lamella extends to 
the basal callus. The basal callus as wide 
as the aperture, porcellanous, rounded, 
slightly raised in mature specimens. Colu- 
mella weakly sigmoid, the basal tooth c^uite 
weak. Suture very slightly impressed. 
Later whorls marked by irregular diagonal 
lines of growth; surface with concentric, 
spiral rows of impressed pits. Protoconch 
IV2 whorls, slightly raised, weakly marked 
by diagonal and regular axial sculpture. 
Periostracum thin; operculum as in genus, 
inner corneous layer light brownish, 
slightly darker at the margins. 



Fleiijht 


Diameter 




mm 


mm 




14.5 


19.5 


San Francisco, near La Tinaja, 
Camagiiey 


13.0 


17.5 


Santa Rita, Camagiiey 


12.5 


19.5 


Paso de las Trincheras, Sierra de 
Cubitas, Camagiiey 


12.5 


16.5 


El Cercado de Cubitas, Camagiiey 



Remarks. This species is very close in 
its relationships to the widely distributed 
suhmarginata, which it resembles in color 
and in the presence of concentric, though 
weak, spiral lines that consist of a series of 
pits. It is easily distinguished by the sharp, 
triangular notch near the posterior (upper) 
angle of the peristome, a characteristic 
which, as Aguayo & Jaume pointed out 
(1954: 10), is found in no other Emoda. 
The color as given above, as well as the 
description of the operculum, is from a lot 
of fresh shells from the Bermudez collec- 
tion labelled merely "Camagiiey." The 
other and better localized shells that were 
available were dead shells, the color hav- 
ing been somewhat faded or stained by 
reddish earth. In all specimens the concen- 
tric spiral lines were quite weak. The 
species apparently is limited to the Sierra 



de Cubitas in the northeastern part of 
Camagiiey Province. This species, as well 
as najazaensis, was misidentified as hasti- 
dana Poey and appeared in the Museo 
Poey and other museum collections under 
this name. Aguayo & Jaume (1954: 11) 
pointed out that hastidana is a mere color 
form of bayamensis, an observation made 
originally by Poey himself (1S65: 144). 

Specimens exam,ined. Camaguey. Las 
Cuevas; Santa Rita v la Entrada del 
Camino de La Guanaja [Santa Rita and the 
entrance to La Guanaja Highway]; Banao; 
Paso de los Burros; Paso de los Trincheras 
y Cueva de los Indios; San Francisco, N of 
La Tinaja; East side of Vereda de Burro, 
Finca San Clemente (USNM); Paso del 
Este (USNM); El Circulo (USNM); Los 
Corrales de Cangilones (USNM); North 
entrance to Paso Escalcra (USNM). All 
of the above localities are in the Sierra de 
Cubitas. 

Emoda najazaensis Aguayo & Jaume 

Plate 1, figures 3, 4; Plate 4, figure 1. 

Emoda najazaensis Aguayo & Jaume 19.54, Tor- 
reia, No. 21, p. 9, pi. l', figs. 7, 9, 10 ("El 
Cacaotal," Sierra de Najaza, Provincia de Cama- 
giiey; holotype, MP 16222: paratypes, MCZ 
257558, from "Vereda del Telegrafo," Sierra del 
Chorillo, ex MP). 

Description. Shell reaching 22 mm in 
diameter, smooth, depressed conic, periph- 
eiy roundly carinate. Whorls 5V4, very 
weakly rounded, body \\'horl descending 
shaiply at the aperture, where it blends 
into the basal callus. Color varied, pale 
olivaceous, pale brown, buff, or faintly 
orange, a pale narrow band at the periph- 
ery. Spire low, mammiform and somewhat 
darker than rest of shell. Aperture widely 
semilunate, lip slightly thickened and 
barely reflected. Columella shallowly con- 
cave below, weakly convex above; basal 
callus weakly lustrous, white, as wide as the 
aperture, rounded at outer margin. Basal 
tooth quite small. Suture moderately im- 
pressed. Spiral sculpture of closely set, 
concentric cords, which are raised in the 



116 Bulletin Muscudi of Comparative Zoology, Vol. 141, No. 3 



pcriostracum, and rcflecttxl below as a line 
of minute impressed pits in the shell, most 
prominent on the body whorl, weaker at 
the base. Protoeoneh IV2 whorls, raised 
nipplelike abo\'e the sneeeeding whorls, 
scnlptur(>d with faint, rc\gular axial striae, 
almost the same eolor as the rest of the 
shell. Periostraeum thin. Opereulum closely 
resembles that of E. bernmdezi, but the in- 
ternal corneous layer is not darker at the 
margin. 



Heijiht 


Diameter 




mm 


mm 




15.5 


20.0 


Vereda del Telegrafo, Sierra del 
Chorrillo, Camagiiey 


14.5 


20.0 


El Cacaotal de Sifonte, Sierra 
Najaza, Camagiiey 


13.0 


19.0 


El Palomar de San Jose, Cama- 
giiey 



Remarks. This species is similar to .stil)- 
iiiar<:,in(ita, especially in color and in the 
presence of spiral concentric incised striae, 
though in najazaensis these are more closely 
spaced. However, it has a generally larger 
shell, which is more depressed and more 
strongly keeled. It appears to be confined 
to the area around the Sierra Najaza 
(Najasa) in the southeastern part of 
Camagiiey Province. The specimens we 
were able to examine, including several lots 
from the type-locality, did not show the 
colors mentioned by Aguayo & Jaume 
(1954: 9). 

Specimens examined. Camaguey. Najaza; 
El Palomar de San Jose; El Cacaotal, 
Najaza; Sierra Guaicanamar; Sitio Afuera; 
Vereda del Telegrafo, Sierra del Chorrillo; 
Sierra del Cochimbo; El Cacaotal de Si- 
fonte, Sierra de Najaza. 

Emoda briarea (Poey) 

Plate 2, figures 1-3; Plate 4, figure 1. 

Helicina hriarca Poey 1851, Memorias Historia 
Natural Isla de Cuba, 1: 108, pi. 11, figs. 9-12 
(San Diego de los Baiios [sic]; lectotype, here 
selected, MCZ 73854, from the J. G. Anthony 
Collection, collected by Gundlach [Trinidad 
Mountains, Santa Clara ( = Las Villas)]. The 
specimen appears originally in the MCZ collection 
as a syntype, identified as such by Torre in June 
1912). 



Helicina briarea Poey, Pfeiffer 1862, Novit. 
Conch., 2: 195, pi. 52, figs. 1-5 (Trinidad). 

Helicina hriaraea "Poey," Sowerby 1866, Thes. 
Conch., 3: 278, pi. 1, figs. 1-3, [error for briarea]. 

Description. Shell very large for the 
family, reaching 34.5 mm in diameter, 
depressed turbinate, solid, roundly carinate, 
sharply so in juvenile specimens and rel- 
atively smooth. Whorls about 5, flat, body 
whorl moderately inflated, flattened ba- 
sally, descending shortly near the aperture. 
Color puqjlish brown on the body whorl 
and somewhat lighter on the spire. There 
is a variously wide, yellowish or whitish 
subsutural band that is extended as a supra- 
peripheral band on the last whorl, where 
it is separated from the light colored basal 
area by a moderately narrow, puiplish red 
band. Subsutural band wanting on early 
postnuclear whorls. Spire depressed, weakly 
rounded except for the slightly raised 
protoeoneh. Aperture widely semilunate, 
evenly rounded, white, ground color and 
spiral band showing through inside. Peri- 
stome expanded, barely flaring above, 
widest just above the periphery and slightly 
thickened. Basal denticle quite small, ob- 
tuse, somewhat lengthened. Parietal callus 
smooth, or yellowish white, more rounded 
than the aperture. Columella obli(iue, 
white, short, concave below. Suture weakly 
impressed. Sculpture of very fine, irregular 
growth lines, without spiral lines. Proto- 
eoneh IV4 whorls, somewhat raised, light 
brown, with faint, curved axial wrinkles. 
Periostraeum thin, yellowish, not carrying 
the shell color, with faint, well-spaced 
spiral ridges. Ojoerculum reddish, darker 
at margins, relatively thin. 



Height 


Diameter 




mm 


mm 




22.5 


34.5 


Puriales, Las Villas 


22.0 


31.5 


Puriales, Las Villas 


18.0 


25.5 


Caracusey, Las Villas 


16.0 


22.0 


San Bias, Las Villas 



Remarks. Most specimens of this species 
can be readily recognized by their large 
size. In pojiulations where the specimens 
are smaller (San Bias, Ciegos de Ponciano, 
etc.), they are determined without diffi- 



Emoda and Glyptemoda in Cuba • Clench and Jacobson 117 



culty by their smooth surface, depressed 
shape, characteristic color, and by the sub- 
sutural band, which has the same color as 
the base. The species differs from the 
larger specimens of sag,raiana from Luis 
Lazo in Pinar del Rio Province in its color 
and in the far less strongly thickened and 
unreflected lip. It differs from E. emoda in 
its more depressed shape, stronger shell, 
and color. The species seems to be con- 
fined to the area around Trinidad in the 
south central part of Las Villas Province. 
As Pfeiffer noted (1862: 196), Poey's lo- 
cality was in error. We here select Trini- 
dad, Las Villas Province, Cuba, as the 
type-locality. Poey selected the trivial 
name referring to the mythological giant 
Briareus, because of the large size of the 
shells. 

Specimens examined. Las Villas. Trini- 
dad: Fomento; Finca Pitajones, Caracusey; 
Santa Rosa, Cafetal Los Puriales; Rio Ca- 
ballero; Portillo, Rio de Ay; Rio Cabunii; 
Ciegos de Ponciano; San Bias Dam above 
San Bias. 

Emoda mayarina mayarina (Poey) 
Plate 2, figure 17; Plate 5. 

Helicina mayarina Poey 1854, Memorias Historia 
Natural Isla de Cuba, 1: 417, pi. 34, figs. 6-8. 
(Mayari [Oriente]; holotype, MP 17034). 

Helicina mayarina Poey, Pfeiffer 1856, Malak. 
BUit., 3: 144; Pfeiffer 1858, Monographia Pneu- 
monopomorum Viventium, Suppl. 1, p. 184; 
Pfeiffer 1862, Novit. Conch., 2: 200, pi. 53, figs. 
6, 7. 

Emoda mayarina (Poey), Aguayo & Jaume 
1954, Torreia, No. 21, p. 7, pi. 2, figs. 1, 6. 

Description. Shell reaching 16 mm in 
diameter, subglobose, solid, moderately 
strong, sublustrous. Whorls 4, convex, the 
last somewhat descending at the aperture 
below the periphery. Color brownish red 
with several faint reddish wrinkles, base 
somewhat yellowish, with a darker periph- 
eral band margined above by a lighter 
band of approximately the same width. 
Spire moderately raised, submammiform. 
Aperture slightly oblique, subsemilunate, 
white at the lip, darker within. Lip simple 



and thin. Parietal callus very thin, white 
and glassy in center, transparent near outer 
margin. Protoconch with fine, cui^ved axial 
striae that grow stronger on the later 
whorls and become vermiculate on the 
entire body whorl, being obsolete only on 
the base and near the aperture, where they 
turn into straight axial lines. Spiral sculp- 
ture wanting. Operculum not seen, but 
presumably like others in this subgenus. 



Height Diameter 
mm mm 



13.0 16.0 Ma>ari, Oriente, (USNM) 

Remarks. This species, according to 
Aguayo & Jaume (1954: 7), is exceedingly 
rare; the Museo Poey has only a single 
specimen. The shells of this species and 
of its two subspecies are easily identifiable 
by their globular shape and especially 
by the variously strong vermiculate or 
wrinkled sculpture. This latter feature 
alone will prevent confusion with any other 
species. 

Aguayo & Jaume believed that the 
nominate species came from the mountains 
of Nipe or Mayari rather than from the 
lowlands around the port of Mayari as 
Poey stated. 

This species was omitted by Wagner in 
his monograph (1907-1908), nor did it 
appear on the list that Fulton (1915) pub- 
lished, citing the species Wagner had 
overlooked. 

Emoda mayarina gufierrezi Aguayo & 
Jaume 

Plate 2, figures 15, 16; Plate 5. 

Emoda mat/arina fiutierrczi Aguayo & Jaume 
1954, Torreia", No. 21, p. 7, pi. 2, figs. 3, 5 
("Finca [Lorenzo Cobos] Estalile," Mayari Arriba, 
Provincia de Oriente; holotype, MP 17342; para- 
type, MCZ 187562). 

Description. This subspecies resembles 
the nominate species in the globular shape 
of the shell, but differs in being larger and 
more strongly sculptured, by having 
stronger spiral ridges on the base, and in 
color, which is predominantly reddish with 
a greenish tinge. 



118 Bulletin Mimitm of Comparative Zoology, Vol. 141, No. 3 



Height Diameter 
mm mm 

15.0 19.5 "Finca Estable," Mayari Arriba 

Remarks. As Aguayo & Jaume comment 
(1954: 8), the color in this form is not of 
specific importance. In this case it is the 
larger size and the more rugose, vermicu- 
late sculptiue which are definitive. 

Specimens examined. Oriente. Finca 
"Lorenzo Cobos Estable," Mayari [Arriba], 
Oriente (E. Pujals). 

Emoda mayarina mirandensis Aguayo & 
Jaume 

Plate 5. 

Emoda mayarina mirandensis Aguayo & Jaume 
1954, Torreia, No. 21, p. 8, pi. 2, figs. 2, 4 
( "Loma de la Cantera" cerca del Central Miranda, 
Palmarito de Canto, Provincia de Oriente; holo- 
type, MP 12593). 

Description. "The present variety is dis- 
tinguished from the typical subspecies by 
its larger size, its somewhat more de- 
pressed form, reddish color and narrower 
and less irregular sculpture" (translated). 

IKiulit Diameter 
mm mm 

16.0 21.0 

Remarks. We have not seen specimens 
of this subspecies. 

Emoda bfanesi Clench & Aguayo 

Plate 1, figures 12, 13; Plate 5. 

Emoda blanesi Clench & Aguayo 1953, in 
Aguayo, Mem. Soc. Cubana Hist. Nat., 21: 301, 
pi. 35, figs. 1, 2 ( "Embarradas" al norte de 
Banes, Oriente; holotype, MP 17324, from Em- 
barradas north of Banes, Oriente, Cuba; paratype, 
MCZ 257553). 

Description. Shell about 18 mm in 
diameter, subglobose, sublustrous and mod- 
erately strong. Whorls 4V4, moderately 
rounded, body whorl subinflated and 
barely descending at the aperture. Color 
reddish, tinged with yellow on the penulti- 
mate and body whorls, the yellow color 
generally becoming stronger as it ap- 
proaches the aperture; peripheral band 
light colored with a darker reddish band 



bordering it below. Spire depressed, 
rounded. Aperture semilunate, white near 
lip, darker inside. Lip thickened inside, 
rounded, not flaring, basal tooth barely 
perceptible. Parietal callus very thin, white 
in the columellar region, transparent near 
outer margins. Suture moderately im- 
pressed. Postnuclear whorls sculptured with 
fine, curved axial cords that grow stronger 
on the successive whorls, the intervals 
being twice as wide as the axial cords on 
the body whorl. These cords are irregular 
and in occasional specimens run together, 
extending to the base without growing 
narrower. Spiral sculpture consisting of a 
subsutural depression on the last whorl, 
and composed of one to three small inci- 
sions on the wider and more depressed axial 
costae. Protoconch IV4 whorls, slightly 
raised and sculptured with fine, curved 
axial wrinkles. Operculum with a thin 
internal corneous layer, yellowish and 
darker at the outer margins. 

Height Diameter 
mm mm 

13.8 16.8 Rio Seco, Banes, Oriente 

13.8 16.2 La Campana, Rio Banes, Oriente 

12.4 17.2 Rio Seco, Banes, Oriente 

Remarks. This species has a limited dis- 
tribution along the north central shore of 
Oriente Province between Banes and 
Gibara. The shell is easily recognized by 
its color, the simple, nonflaring lip, the 
subglobose shape, and the impressed sub- 
sutural line. It differs from pulcherrima in 
having stronger axial costae, a simple lip, 
and in the presence of the subsutural 
groove with the incisions on the summit 
of the axial cords. 

Specimens examined. Oriente. La Cam- 
pana, Rio Banes; Rio Seco, Banes; Antilla; 
Loma de Carbon, barrio de Rio Seco; N 
of Banes (both USNM). 

Genus Glypfemoda Clench & Aguayo 

Glyptcmoda Clench & Aguayo 1950, Rev. Soc. 
Malacol. 'Carlos de la Torre,' 7: 61 (as a sulv 
genus; type-species, Helicina torrci Henderson 
1909, original designation). 



Emoda and Glyptemoda in Cuba • Clench and Jacobson 119 



Description. Shell like that of Emoda 
sensu stricfo, but more depressed, exceed- 
ingly rough, dirty white in color, occasion- 
ally tinged with lemon yellow or purple, 
and sculptured with strong, thin lirae, 
separated by flat, much wider intervals. 
The operculum has the outer surface 
covered with small, more or less equisized, 
rounded, glassy granules. In this character 
it is between the operculum of Emoda 
with its unmarked surface and the strongly 
roughened surface of the operculum of 
ScJiasiclteila. Periostracum wanting. 

Remarks. Keen (1960: 286) synony- 
mized Glyptemoda with Emoda s. s. We 
cannot agree with this decision. Instead 
we feel that this taxon should be raised to 
full generic rank. 

Baker (1922) showed that the lateral 
tooth complex in Helicininae consists of a 
hammer-shaped comb-lateral structure with 
the basal pillar located at one end. In the 
subfamily Proserpininae, the pillar is cen- 
trally located, giving an anvil or T-shape 
to the structure. In Glyptemoda the basal 
pillar, consisting of two united portions 
separated by a deep vertical furrow, is 
almost as large as the top of the tooth itself 
(PI. 7, fig. 1). Apparently this type of 
pillar can be derixed more easily from 
Proserpininae than Helicininae but may be 
distinct enough to stand in a subfamily 
itself. There are other radular peculiarities 
in Glyptemoda. Five of the seven central 
teeth are oblong in shape and are weakly 
cusped. In this respect they differ from 
the lanceolate or obliquely lanceolate cor- 
responding teeth of the Helicininae and 
Proseipininae. The numerous uncinals are 
strongly curved and have one or two \\'eak 
cusps on the inner marginals to three to 
fi\'e rather strong cusps on the outer ones. 

These radular characteristics, together 
with the strong surface sculpture and de- 
pressed shape of the shell, as well as the 
granule-covered outer surface of the oper- 
culum, induce us to recognize Glyptemoda 
as a full genus. 



Glypfemodo forrei torrei (Henderson) 

Plate 3, figures 1, 2; Plate 5. 

HeJicina torrei Henderson 1909, Nautilus, 23: 
50, pi. 4, figs. 1-3 (Los Negros, 25 miles [sic] 
southeast of Bayanio in the province of Oriente, 
Cuba, "in woods on low limestone lulls"; holotype, 
USNM 463791, paratypes, MCZ 20863). 

Description. Shell much depressed, reach- 
ing almost 30 mm in diameter, lustre- 
less, rough and veiy solid. Whorls 4y4, 
rapidly widening, almost flat, the last 
whorl descending shortly at the aperture. 
Color grayish white or cream, occasional 
specimens sho\\'ing yellow or reddish 
purple on the spire. Spire planiform or 
slightly raised. Aperture semilunate, evenly 
rounded, generally white inside, occasional 
specimens with a tinge of yellow deep 
within. Lip flaring, white, thickened x\'ithin, 
reflected aboxe, widest above, gradually 
narrowing at the base, where it terminates 
in a very small, rounded tubercle. Parietal 
callus white, lustrous, not raised at the 
outer edge. Columella white, short, convex 
above, almost straight below. Suture well 
impressed. Surface with 17 to 20 thin, 
sharply raised spiral cords, crossed by 
regular, rather strong, diagonal axial sculp- 
ture. This sculpture crenulates the upper 
edge of the spiral cords. Intervals much 
wider than the cords, crossed by the axial 
growth lines. Protoconch 2 large whorls, 
smooth but with closely spaced, cun'cd 
axial wrinkles, rather more closely spaced 
than in most species of Emoda sensu 
stricto. No periostracum. Operculum as 
in subgenus. 

Height Diameter 
mm mm 

16.5 28.9 Los Negros, Jiguani, Oriente, para- 
t>'pe 

16.0 29.0 Los Negros, Jiguani, Oriente, para- 
type 

14.0 30.5 Los Negros, Jiguani, Oriente, para- 
type 

13.5 26.0 Los Negros, Jiguani, Oriente, para- 
type 

Remarks. The number of spiral ridges 
varies from 17 to 20 in a series of paratype 



120 Bulletin Museum of Comparative Zoology, Vol. 141, No. 3 



specimens and from 15 to 20 in a series 
from the de la Torre collection merely 
labelled "Los Negros, Jiguani." This vari- 
ation in number may be due to the pres- 
ence of thinner and lower subsidiary ridges 
in the intervals in some specimens. The 
colors as given by Henderson do appear 
in isolated specimens, but most, even those 
taken alive, are white or cream colored 
under an encrusting layer of dirt. The 
spire, which is completely flat in some 
specimens, is moderately to well raised in 
others. This might be a secondary sexual 
characteristic. Henderson's location of Los 
Negros at 25 miles southeast of Bavamo is 
in error. Los Negros in Oriente is only 
about nine miles from Bayamo. 

Specimens examined. Oriente. Los Ne- 
gros, 15 km SE of Bayamo. 

Glyptemoda forrei freirei Clench & Aguayo 
Plate 3, figures 3, 4; Plate 5. 

Emoda {Glijptemoda) ionci freirei Clench & 
Aguayo 1950 Rev. Soc. Malacol. 'Carlos de la 
Torre,' 7: 61, pi. 12, figs. 1-3 (La Manuela, 
antes de llegar a la Furnia de la Mina Los 
Negros, Baire, Oriente; holotype, MP 12918; para- 
types, MCZ 18.5798). 

Emoda pujaisi .\guayo & Jaume 1954, Torreia, 
No. 21, p. 5, pi. 1, fig. 8 ("La Cloria," Union 
del Camino de Maffo a San Lorenzo y a Los 
Negros, Jiguani, Provineia de Oriente; holotype, 
MP 17386). 

Description. Shell as in torrei but differ- 
ing in its smaller diameter (5 to 8 mm less 
than the nominate species), its proportion- 
ately more elevated shape, and by the 
spiral ridges, which are lower and flattened 
at the top rather than sharp. The speci- 
mens from the Cueva de Pepin are tinged 
with yellow or brick red. 

Height Diameter 
mm mm 

14.5 20.0 "La Gloria," Maffo (as E. puialsi) 
14.3 20.0 "La Gloria," Maffo (as E. pujaisi) 
14.0 21.5 La Manuela (type-locality) 

Remarks. According to Clench & Aguayo, 
Jose A. Freire, who collected the type 
specimens, reported that he found this sub- 



species in the lowlands, whereas torrei 
torrei lives in the higher elevations. We 
have not seen the operculum, but presum- 
ably it is similar to that of torrei torrei. 

Emoda pujaisi Aguayo & Jaume is a 
neighboring population that differs from 
the present subspecies only in degree. The 
two forms are found close together in the 
vicinity of Los Negros, and there is little 
doubt that they are the same. 

Specimens examined. Oriente. La Muela; 
Cueva de Pepin, Baire (USNM); "La 
Gloria," Maffo, Jiguani (type locality of 
E. pujaisi). 

Emoda (?) dementis Clench & Aguayo 

Plate 3, figure 5. 

Emoda dementis Clench & Aguayo 1950, Re\-. 
Soc. Malacol. 'Carlos de la Torre,' 7: 62, pi. 12, 
figs. 1, 5; (Cayo del Key, Mayari, Oriente; holo- 
type, MP 12923; paratype, MCZ 185799). 

Description. Shell imperforate, thin, 
fragile, and translucid. Color uniformly 
bluish green. Whorls 2 in the juvenile 
specimens, suture well impressed, body 
whorl with rounded keel. Spire depressed, 
little raised above the plane of the body 
whorl. Aperture semilunate; umbilical cal- 
lus microscopically granulose. Columella 
short, blending gradually into the basal lip. 
The spiral sculpture' consists of irregular 
microscopic striae grouped in unef(ual 
series; axial sculpture of fine irregular, 
diagonal, growth lines. Protoconch P/j 
whorls, axially striated, the striations 
crossed by spiral lines; with four curved, 
rounded furrows at the distal portion. 
Periostracum imperceptible; opercukun 
thin, green, smooth, nucleus submarginal, 
with a sculpture of concentric lines. 



Height Diameter 
mm mm 

6.5 10 



(juvenile) 



Remarks. This species was described on 
the basis of four juvenile specimens, and 
was doubtfully assigned to Emoda. These 
doubts are justified. The protoconch, as 
described bv the authors and as seen in the 



Emoda and Glyptemoda in Cuba • Clench and Jacobson 121 



specimen in MCZ, lacks the regular, curved, 
axial striae or wrinkles that characterize 
the genus, and also shows faint spiral 
sculpture that is not found on the proto- 
conch of Emoda sensu stricto. The shell is 
very fragile, almost transparent, weaker 
than the customarily strong shell of Emoda, 
even in juvenile specimens. The true place- 
ment of dementis Clench & Aguayo must 
await the discovery of additional and more 
mature specimens. 

REFERENCES CITED 

Adams, H. and A. 1856. The Genera of Recent 
Mollusca, London, 1: 300-309, 3: pi. 87, 
(Helicinidae). 

Aguayo, C. G., and M. L. Jaume. 1954. Notas 
sobre el genero Emoda, Torreia, Habana, 
no. 21: 4-13. 

Arango, R. 1878-1880. Contiibucion a la Faima 
Malacologica Cubana, Havana, pp. 280 + 35 
(inde.x and errata). 

Baker, H. B. 1922. Notes on the radula of the 
Helicinidae, Proc. Acad. Nat. Sci. Philadel- 
phia, 74: 29-67. 

. 1926. Anatomical notes on American 

Helicinidae, ibid., 78: 35-56. 

1940. New snbgenera of Antillean Heli- 



cinidae. Nautilns. 54: 70-71. 

Bourne, G. C. 1911. Contribntions to the mor- 
phology of the gronp Neritacea at the aspido- 
branch gastropods. Proc. Zool. Soc. London, 
1911: 759-809. 

Clench, William J., and Morris K. Jacobson. 
1968. Monograph of the Cuban genus Viana 
(Mollusca: Archaeogastropoda: Helicinidae), 
Breviora, Mus. Comp. Zool., no. 298: 1-25. 

. 1970. The genus Priotrochatella of the 

Isle of Pines and Jamaica, West Indies. Occas. 
Pap. Moll., Mus. Comp. Zool., Cambridge, 
Mass., 3(39): 61-80. 

Dance, S. P. 1966. Shell Collecting. Berkeley 
and Los Angeles: University of California 
Press, 344 pp. 

Fischer, P. 1885. Manuel de Conchyliologie, 
Paris, pp. 794-799, figs. 554-557 (Helicini- 
dae). 

FoRCART, L. 1950. Von Dr. H. P. Schaub auf der 
Insel Cuba gesammelte Land- und Siisswasser- 
schnecken. Arch. Molluskenk. 79: 67-72. 

Fulton, H. C. 1915. On Dr. Anton Wagner's 
Monograph of Helicinidae in the Conchylien- 



Cabinet, 1911. Proc. Malac. Soc. London, 

11: 237-241; 324-326. 
Isenkrahe, C. 1867. Anatomic von Helicina 

titanica. Arch. Naturg., 33rd year. Vol. 1, 

pp. 50-72. 
Keen, M. 1960. Helicinidae. In Treatise on In- 
vertebrate Paleontology, Mollusca, University 

of Kansas, Lawrence, Kansas, 1 : 285-288, 

figs. 186-187. 
Lea, I. 1834a. Observations on Naiades and 

Descriptions of New Species of that and 

other families. Trans. American Phil. Soc. 

(N. S.), 5: 23-119. 
. 1834b. Observations on the genus Uiiio, 

Philadelphia, 1- 1-232. 
Mayr, E. 1969. Principles of Systematic Zoology. 

New York: McGraw-Hill, 11 + 428 pp. 
Orbigny, a. d'. 1842. Mollusques. In Sagra, 

Histoire Physique, Politique et Naturelle de 

rile de Cuba, Paris, 2: 1-264. 
Pfeiffer, L. 1850-1853. Helicinacea. In Martini 

& Chemnitz, Conchylien-Cabinet, ( 2 ) 1 : 

sect. 13, pt. 1, 78 pp. 
. 1854-1860. Novitates Conchologicae, 

Cassel, 1: 1-138; ibid., 1860-1866, 2: 140- 

303. 

1860. Zur Molluskenfauna der Insel Cuba, 



Malak. Bliit, 6: 66-102. 

PoEY, F. 1851-1854. Memorias sobre la Historia 
Natural de la Isla de Cuba, Havana, 1 : 
1-463; op. cit., 1856-1866, 2: 1-442. 

: — . 1865. Notas. In Rafael Arango, Catalogo 

de los Moluscos Terrestres y Fluviales de la 
Isla de Cuba, Repertorio fisico-natural de la 
Isla de Cuba, Havana, pp. 71-149. 

Reeve, L. a. 1874. Conchologia Iconica, London, 
Vol. 19, 34 pis. and letter press (Helicini- 
dae ) . 

SowERBY, G. B., II. 1847. Thesaurus Conchy- 
liorum, London, Vol. 1, monograph of the 
genus Helicina, pp. 1-16. 

. 1866. O)). fit., Vol. 3, second monograph 

of the genus Helicina, pp. 277-302. 

Thiele, J. 1929. Handbuch der systematischen 
Weichtierkunde, Jena, 1: 80-91, figs. 57-67, 
( Helicinidae ) . 

Troschel, F. H. 1857. Das Gebiss der Schnecken, 
Berhn, 1: 75-85 (Helicinidae). 

Wagner, A. 1907-1908. Helicinidae. In Martini 
& Chemnitz, Conchylien-Cabinet, Niirnberg, 
(2) 1: sect. 18, pt. 2, pp. 46-138. 

Wenz, W. 1938. Gastropoda. In Handbuch der 
Paliiozoologie, Helicinidae, Berlin, 6: pt. 1, 
pp. 435-448. 

(Received 15 January 1970.) 



122 Bulletin Museum of Comparative Zoology, Vol. 141, No. 3 



PLATE 1. Figs. 1 and 2. Emoda submarginata (Gray), El Grille, Madruga, Havana, MCZ 60969, Topofype (1.5 X). Figs. 3 
and 4. f. na;ozaens/s Aguayo & Jaume, El Cacaotal, Sierra de Najaza, CamagiJey, MCZ 257422, Paratypes (1.5 X)- Figs. 
5 and 6. E. bermudezi Aguayo & Jaume, Banao, Cubitas Mountains, Camaguey, MCZ 262719 (1.5 X)- Fig. 7. £. ciliata 
(Poey), San Juan de Letran, Trinidad, Las Villas, Lectotype, MCZ 262717 (1.5 XI- Figs. 8 and 9. £. ciliata (Poey) same 
localify, MCZ 127230, Topotypes (1.5 Xl- Figs. 10 and 11. E. ciliata gu/sono Wagner, 'Cuba," MCZ 262718 (1.5 Xl- 
Figs. 12 and 13. E. b/ones/ Clench & Aguayo, Rio Seco, Bones, Orients, MCZ 1 35961 (1.2X1- Fig. 14. E. pu/c/ierrimo (Lea), 
USNM 104613, Lectotype (1.5 Xl- Fig. 15. E. pulcherrima (Lea), Trans. American Philos. Soc. (NS), pi. 19, fig. 57, photo- 
graphic copy. Figs. 16 and 17. E. sagraiana (Orbigny), Ensenada de los Burros, Cabezas, Pinar del Rio, MCZ 127148 (1.5 
XI- Figs. 18 and 19. E. sograiono forma percrassa Aguayo & Jaume, La Gijira, Luis Lazo, Pinar del Rio, MCZ 127253 (1.5 
XI- Figs. 20 and 21. E. silacea (Morelet), Yunque de Baracoa, Oriente, Topotype, MCZ 127229. 



Emoda and Glyptemoda in Cuba • Clench and Jacobson 123 




124 Bulletin Museum of Comparative Zoology, Vol. 141, No. 3 




PLATE 2. Fig. 1. Helicina briarea Poey, Trinidad Mountains, Santa Clara [Las Villas], MCZ 73854, Lectotype (1.4 X)- 
Figs. 2 and 3. E. briarea (Poey), "Puriales," Trinidad Mountains, Las Villas, MCZ 58560 (1.4 X)- Fig- 4. Helicina baya- 
memii Poey (Buena Vista, Bayamo, Oriente) MCZ 73855, Lectotype (c. 1.4 X)- Fig. 5. ditto, MCZ 256496, Paralectotype. 
Figs. 6 and 7. Emoda emodo Pfeiffer, Manatial de Santa Fe, Oriente, MCZ (1.2 X). Figs. 8 and 9. He//c/no titanica Poey, 
Barocoa (Oriente), MCZ 256495, Syntypes (1.2 X)- Fig. 10. Emodo pujalii Aguayo & Jaume (^ torre; treirei Clench & 
Aguayo), Loma La Gloria, Maffo, JiguanI, Oriente, MCZ 263902, Topotype (2.1 X). Fig, 11. Emoda caledonlensis n. sp.. 
Mine Caledonia, Mayari, Oriente, MCZ 261352, Paratype (1.2 Xl- Fig. 12. ditto, 1 km from pine forest, Mayari, MCZ 
261349, Paratype (1.2 X)- Figs. 13 and 14. E. pulcherrima (Lea), Pico Turquino, bet. 2000 and 5000 ft., Oriente, MCZ 
267577 (1.5 X). Figs. 15 and 16. £. mayan'na gutierrezi Aguayo & Jaume, Finca Estable, Mayari Arriba, Oriente, Paratype, 
MCZ 187562 (1.2 X)- Fig. 17. Helicina mayarina Poey, Mayari, Oriente, Syntype, USNM. (1.5 X)- 



Emoda and Glyptemoda in Cuba • Clench and Jacohson 125 




PLATE 3. Figs. 1 and 2.^ Glyptemoda torrei torrei Henderson, 
Los Negros, Jiguani, Oriente, MCZ 20863, Paratypes (1.5 X). 
Figs. 3 and 4. Glyptemoda torrei freirei Clench & Aguayo, 
La Manuela near the mine, Boire, Oriente, MCZ 185798, 
Poratype (1.5 X)- Fig. 5. fmodo ? dementis Clench & 
Aguayo, photographic copy of original figure. Rev. Soc. 
Mai., 7: pi. 12, fig. 4. Figs. 6 and 7. Helicina crassa 
Orbigny ( = pukherrima Lea), Syntype, BM(NH) No. 
1854.100.4.169 (1 X). (Photo courtesy of Angela Cane, 
BM(NH).) 



126 Bulletin Museum of Comparative Zoology, Vol. 141, No. 3 




fig.i 




PLATE 4. Fig. 1. Distribution of (1) Emoda sagraiana (Orbigny); (2) E. briarea (Poey); (3) E. ciliata (Poey); (4) E. bermudezi 
Aguayo & Joume; (5) E. najazaensis Aguayo & Jaume. Fig. 2. Distribution of Emoda submarg/nafo (Gray). 



Emoda and Glyptemoda in Cuba • Clench and Jacohson 127 




o 

c 



°=, E 
o 



Uj 



<u 



i? oa 



CD 
< 



oa 



r5 t U 

to (U 

^' t ^ 

U D -i 

c ._ 

TT • '^ 

D Uj 

u . — , 






o — 



■ o 

Lu U 

O 

CO "" 

oa 



u 



D 
O 

E 



(U 



-^ S O 



c 
a"! 





"o 
o 

E 

Uj 



o 

D 



-i; oa 

^ O u 



<u 



c >~ 






D _ — 

O dj" 

E j; E 

^ >^ => 

° ^ ^ 

c — oa 

• 2 D O 

J I r 

i O) O) 

- Z) c 

IT) «/i <D 



^ E 



128 BuUetin Museum of Comparative Zoology, Vol. 141, No. 3 




PLATE 6. 1. Digestive gland. 2. Right columellar retractor muscle. 3. Male reproductive gland. 4. Intestine. 5. Pos- 
terior end of mantle cavity. 6. Nephripore. 7. Pericardium. 8. Left columellar muscle. 9. Gonads. 10. Operculum. 
11. Base of foot. 



Emoda and Glyptemoda in Cuba • Clench and Jacobson 129 




PLATE 7. Fig. 1. Glyptemoda torrei (Henderson), Paratype, 
MCZ 20863, Los Negros, 25 mi. SE of Boyamo, Oriente, 
Cuba. Central and lateral tooth complexes, showing main 
plate of lateral complex only. Fig. 2. Emoda submarginata 
(Gray), MCZ 127233, El Mamey, Buena Vista, Las Villas, 
Cuba, ex Bermudez. Central and lateral tooth complexes, 
showing main plate of lateral complex only. 



Table 1. Distribution of the Species of 
Emoda and Glyptemoda 



-a 

u 
a 
a 



G 
a 
> 

a 

X 



sagraiana 

silacea 

pulcherrima 

p. titanica 

suhmarginata 

ciliatu 

c. guisana 

enwda 

caledonietisis 

haijamensis 

bermtidezi 

Jiajazaetisis 

hriarea 

mcujarina 

m. gutierrezi 

m. mirandensis 

hlanesi 

torrei torrei 

torrei freirei 



X 



X 



X 



X 



> 
1-1 



X 
X 

? 



>, 

:3 

us 



o 



X 



X 
X 



•c 

o 



X 
X 
X 
X 



X 
X 
X 



X 
X 
X 
X 
X 
X 



130 Bulletin Museum of Conipamtive Zoulugij, Vol. 141, No. 3 



INDEX 

Alcadia, 99, 100, 102, 103 

ba^i^eiisis, Alcadia nuda, 102 
hastidaiui, llcUcina, 113, 111, 115 
basUdknia, IlcUcina, 113 
bayiunensis, Enioda, 113, 114, 115 
bayhamensis, Helicina, 113 
bennndezi, Emoda, 100, 110, 114 
blanesi, Emoda, llcS 
hriaraea, Helicnia, 1 Hi 
briarea, Emoda, 101, 112. 116 
hrkirca. Ilclicina, 116 

caledonit'iisis, Emoda, 99, 112 

cat (dine mis, Helicina, 105 

catuUnianu, Ilclicina, 105 

Ceratodiscus, 99 

ciliiita, Alcadia (Emoda), 110 

ciliata, Emoda, 100, 102, 103, 110, 111, 114 

ciliiita, Helicina, 1 10 

dementis, Emoda, 120, 121 

crassa, Helicina, 101, 107, 108 

Emoda. 99, 100, 101, 102, 119, 120, 121 
emoda, Alcadia (Emoda) saf:iaiana, 112 
emoda, Emoda, 112, 117 
emoda, Emoda sagraiana, 112 
emoda, Helicina, 112 
Eutiochatella, 99 

fossuUita, Helicina, 110, 111 

freirei, Emoda (Glyptemoda) toirci, 120 

Ircirei, (dypteiiioda torrei, 120 

Glyptemoda, 99, 118, 119 
gtii-sana, Alcadia (Emoda) ciliata, HI 
miisana, Emoda ciliata. 111 
miticrrezi, Emoda mayariiia, 117 

Helicina, 99 
Hclicinidac, 99 
llelicininae, 119 

jeannercii, Helicina, 113, 114 

lamellosus, Thais, 106 



lapillus, Thais, 106 
Liuidclla, 99 

mayarina, Emoda mayarina, 117 
ntaijarina, Helicina, 117 
niirandensis, Emoda mayarina, 118 

najazaensis, Emoda, 100, 110, 115, 116 
nuda, Alcadia, 102 



OL 



hracca, Helicina, 100, 104, 105 



])erc>a.ssa, Emoda sagraiana, 100, 105, 106 

phnospira, Alcadia (Emoda) ptdcherrima, 109, 110 

Priotiochatella, 99 

Proserpina, 99 

Proserpininae, 119 

pulcherrima, Emoda pulchcrrima, 100, 107, 108, 

109, 110, 114, 118 
pulchcrrima, Helicina, 99, 107 

rubra, Helicina, 109 
rubrocinctu, Helicina, 107, 108 

sagra, Helicina, 105, 114 

sagra, Trochatclla, 105 

sagraiana, Emoda, 100, 102, 103, 105, 112, 114, 

117 
sagraiana, Helicina, 105, 106 
Schasicheila, 100, 101, 102, 119 
silacea, Alcadia (Emoda), 104 
silacea, Emoda, 101, 102, 104, 111 
Striatemoda, 101 
striatura, Alcadia, 101 
submarginata, Emoda, 100, 103, 108, 109, 110, 

111, 114, 115, 116 
submarginata, Helicina, 109 

titanica, Alcadia ( Emoda ) ptdcherrima, 108 

titanica, Emoda pulchcrrima, 103, 108, 113, 114 

titanica, Helicina, 103, 108 

torrei, CUyptemoda torrei, 119, 120 

torrei, Helicina, 119 

Viana, 99, 100 

zaya.si. Emoda, 109 



Note: Synonyms are italicized. 






OF THE 



Museum of 

Comparative 

Zoology 



The Diadematus Group of the Orb- Weaver 

Genus Araneus North of Mexico 

(Araneae: Araneidae] 



HERBERT W. LEVI 



HARVARD UNIVERSITY VOLUME 141, NUMBER 4 

CAMBRIDGE, MASSACHUSETTS, U.S.A. 4 FEBRUARY 1971 



PUBLICATIONS ISSUED 

OR DISTRIBUTED BY THE 

MUSEUM OF COMPARATIVE ZOOLOGY 

HARVARD UNIVERSITY 

Bulletin 1863- 

Breviora 1952- 

Memoirs 1864-1938 

JoHNSONiA, Department of Mollusks, 1941- 

OccAsioNAL Papers on Mollusks, 1945- 

Other Publications. 

Bigelow, H. B., and W. C. Schroeder, 1953. Fishes of the Gulf of Maine. 
Reprint, $6.50 cloth. 

Brues, C. T., A. L. Melander, and F. M. Carpenter, 1954. Classification of 
Insects. $9.00 cloth. 

Creighton, W. S., 1950. The Ants of North America. Reprint, $10.00 cloth. 

Lyman, C. P., and A. R. Dawe (eds.), 1960. Symposium on Natural Mam- 
malian Hibernation. $3.00 paper, $4.50 cloth. 

Peters' Check-list of Birds of the World, vols. 2-7, 9, 10, 12-15. (Price list 
on request. ) 

Turner, R. D., 1966. A Survey and Illustrated Catalogue of the Teredinidae 
(Mollusca: Bivalvia). $8.00 cloth. 

Whittington, H. B., and W. D. I. Rolfe (eds.), 1963. Phylogeny and Evolution 
of Crustacea. $6.75 cloth. 

Proceedings of the New England Zoological Club 1899-1948. ( Complete sets 
only. ) 

Publications of the Boston Society of Natural History. 



Authors preparing manuscripts for the Btilletin of the Museum of Comparative 
Zoology or Breviora should send for the current Information and Instruction Sheet, 
available from Mrs. Penelope Lasnik, Editor, Publications Office, Museum of 
Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138, 
U.S.A. 



© The President and Fellows of Harvard College 1971. 



THE DIADEMATUS GROUP OF THE ORB-WEAVER GENUS 
ARANEUS NORTH OF MEXICO (ARANEAE: ARANEIDAE) 



HERBERT W. LEVI 



ABSTRACT 

Among our commonest spiders are the large 
Aiancus orb-weavers. Eighteen species are found 
north of Mexico. They are redescrilied, their 
diagnostic characters ilhistrated, their ranges 
mapped, and notes on their habits recorded. Three 
species turned out to be new: A. loashingtoni from 
northeastern U. S. and eastern Canada; A. yiikon 
from Yukon Territory; and A. pima from the 
Southwest. 

INTRODUCTION 

Among the species included in the 
present revision are some of the commonest 
spiders of North America, although several 
Araneus species are quite rare. Collections 
were examined not only to get an idea of 
the variation and distribution of common 
species, but also with the hope of finding 
members of the rarer species. The enormity 
of the collections available slowed down 
the work immensely. 

Because of the urgent need for identifi- 
cation and information on the distribution 
of the common spiders, I am presenting 
separately the diademotus group of the 
genus Aroneus, some members of which 
have never been illustrated before. It may 
take many years before I have examined 
types of all names of the numerous genera 
placed within the family and the many 
species described in the genus Aroneus in 
the Americas. At present I do not know 
the limits of the genus and have not de- 
cided whether certain species are best 
included in the genus Araneus. 



The recognition of reliable diagnostic 
morphological features of species was one 
of the critical questions to be resolved. My 
conclusions agree entirely with those of 
Grasshoff (1968), but differ from those 
published by Archer (1951a, b). Archer 
tliought that the species can be diagnosed 
by the shape of one single sclerite of the 
male palpus, the median apophysis, and 
the ventral view of the scape and epigynum 
of the female. Perhaps these structures will 
prove of value in differentiating genera; 
they certainly are of quite limited value 
for species diagnosis. Also, I agree with 
Grasshoff that leg spines have doubtful 
value in separating males of Araneus spe- 
cies; they are extremely variable within 
populations (L. D. Carmichael, in manu- 
script ) . 

As in my revisions of theridiid spiders, 
long established names of common species 
were kept; changing them does not make 
sense. ^ 



^ I am following the purpose of the International 
Code on Zoological Nomenclature as expressed in 
its Preamble, although other authors have oc- 
casionally in my opinion intei-preted individual 
pro\isions out of the context to obligate tlie 
changing of names. For many common North 
American species older names are available, but 
these often have doubtful application as the types 
have been lost, and interpretation of the usually 
inadequate description depends on the experience 
of the reader. Doujjtful also are many of the 
names used bv Chamberlin and Ivie (1944) in 



Bull. Mas. Conip. Zool, 141(4): 131-179, February, 1971 131 



132 Bulletin Mmeum of Comparative Zoology, Vol. 141, No. 4 



In this study the work of the late Dr. 
H. Wiehle on European spiders has been 
of tremendous help, as has also the recent 
study on variation and moiphological cri- 
teria of several European Amneus species 
by Grasshoff (1968). 

I wish to thank the numerous biologists 
who have contributed specimens for study: 
Mr. J. D. Berman; Mr. D. Bixler; Dr. R. E. 
Crabill of the United States National 
Museum; Mr. Bruce Cutler; Dr. C. D. 
Dondale sent collections of the Canadian 
National Museum; Dr. M. G. Emsley of the 
Academy of Natural Sciences, Philadelphia; 
Dr. W. J. Gertsch of the American Museum 
of Natural History; Dr. O. Kraus of the 
Senckenbergische Naturforschende Gesell- 
schaft; Mr. Robin Leech; Mr. G. H. Locket; 
Dr. D. C. Lowrie; Dr. M. Moritz of the 
Zoologisches Museum der Humboldt Uni- 
versitat, Berlin; Mr. P. L Pcrsson of the 



Natural History Museum, Stockholm; Mr. 
T. R. Renault of the Canada Department 
of Forestry, Fredricton, New Brunswick; 
Miss Susan Riechert; Mr. V. D. Roth of 
the Southwestern Research Station; Dr. 
J. G. Sheals and Mr. D. Clark of the 
British Museum (Natural History); Dr. R. 
Snetsinger; Mr. W. A. Shear; Dr. E. Sutter 
of the Naturhistorisches Museum, Basel; 
Dr. S. L. Tuxen and Mr. B. Petersen of the 
Universitetets Zoologiska Museum, Copen- 
hagen; Mr. L Valovirta, University of Hel- 
sinki Zoological Museum; Dr. H. V. Weems, 
Jr. and Mr. K. J. Stone of the Florida Col- 
lection of Arthropods; Dr. T. Yaginuma; 
and Dr. G. Edmundson and the staff of 
the University of Utah collections. My wife 
has helped with writing and editing. 

This investigation was su]Oported in part 
by Public Health Service Research Grant 
AL01944 from the National Institutes of 
Allergy and Infectious Diseases. 



their attempt to resurrect names of Walckenaer 
for the manuscript ilkistrations of Abbot. Sub- 
jective judgment has to be used in interpreting 
tliese rather primitive ilkistrations of the dorsal 
aspect of spiders, tlie types of these names. While 
other biologists may feel sure of their judgment 
in these dou])tful cases, I am not, and after hav- 
ing obtained copies of Abbot's drawings I find 
that I disagree with many of the Chamberlin 
and Ivie assignments; many are obviously in error 
(Levi and Levi, 1961). The Statute of Limi- 
tations ( Art. 23b ) permits us now to assign many 
of these uncertain ancient names to the status of 
obsolete names, should they threaten familiar 
names that have been in common use for more 
than 50 years. Like any other tool used by the 
taxonomist, the Code has to be used with common 
sense. 

My request to the Commission to resolve by 
plenary power the problem of the two family 
names, Araneidae versus Argiopidae, has been 
withdrawn because only a few, often those in 
opposition and those with least experience in 
taxonomy and nomenclature, state their opinions 
in writing to the Secretary. However, the non- 
controversial request to place the name Argiope on 
tlie Official List of Generic Names in Zoology has 
not been witlidrawn. Although sent to the Secre- 
tary of the Commission in May 1967, it has not 
been printed yet. Once printed the name is 



protected by Art. 80 of tlie Int. Code Zool. 
Nomcncl. 

Instead of submitting the problem of the 
family name to the Commission, I have circulated 
a request for the opinion of colleagues. The poll 
included, of course, nontaxonomists working witli 
orb- weavers. A majority of correspondents favored 
Araneidae (Araneidae, 43; Argiopidae, 29; Argyo- 
pidae, 1; outside of North America, Araneidae, 
28; Argiopidae, 25; Argyopidae, 1). I will con- 
tinue to use the name Araneidae. 

The problem of the spelling of spider generic 
names is puzzling. The Code (Art. 32) now 
considers incorrect the emendations of spellings 
by Thorell, almost universally adopted for 100 
years. Changing the names now would conflict 
witli Art. 23b (since the incorrect emendations 
are junior objective synonyms. Art. 33a), and also 
with the purposes of the Code and would cer- 
tainly be wrong. Thus it seems to me that we 
have to follow the long accepted spellings of 
generic names as also generally adopted by 
Bonnet in the Bihliographia Arauconim, and at- 
tempt to bring critical cases to the attention of 
the Commission. It is possible tliat the pertinent 
paragraphs of the Code might be clarified by a 
future International Congress to a\oid unneces- 
sary name changes. 



Araneus diadematus group Orb-weavers • Levi 133 



Araneus Clerck^ 

Araneus Clerck, 1757, Svenska Spindlar, p. 15. 
Type species designated by Simon, 1893, His- 
toire Naturelle des Araignees, 1: 829, A. angu- 
latiis Clerck. But the type is said to be A. 
diadematus Clerck by Petrunkevitch, 1928, 
Trans. Connecticut Acad. Sci., 29: 136 and 
Bonnet, 1955, Bibliographia Araneorum, 2: 408, 
altliough the type species is correctly stated to 
be A. angtdattis by Petrunkevitch, 1911, Bull. 
Amer. Mus. Natur. Hist., 29: 255 and by 
Bonnet, 1950, Bull. Soc. d'Hist. Natur. Toulouse, 
85: 1-9. 

Aranea Linnaeus, 1758, Systenia Naturae, 10th ed. 
p. 619. The only genus of spiders; A. diadema 
is listed first. No valid type designation seems 
to have been made previously, thus I here 
designate A. diadema Linnaeus. The generic 
name Aranea has always assumed to be a 
synonym for Araneus Clerck. 

Epeira Walckenaer, 1805, Tableau des Araneides, 
p. 53. Type species designated by Latreille, 
1810, Considerations Generales, p. 424, Aranea 
diadema Linnaeus; a second designation is by 
Thorell, 1869, On European Spiders, p. 53, 
Epeira diademata (Clerck). In 1928 Petrunke- 
vitch (Trans. Connecticut Acad. Sci., p. 136) 
indicated that E. cornuta (Linnaeus) [sic] was 
the type. Presumably he meant Araneus cor- 
nutus Clerck. 

Neopora Simon, 1864, Histoire Naturelle des 
Araignees, p. 261. A name for a subgenus. The 
type species is N. diadema ( = Araneus diade- 
matus Clerck) designated by Bonnet, 1958, 
Bibliographia Araneorum, 2: 3054. 

Burgessia McCook, 1894, American Spiders, 3: 
182. A subgenus for the group comprising 
corticaria, miniata, honsallae, mayo, hispinosa, 
paeificae as well as forata, linteata, and juniperi. 
The type species is Epeira corticaria (Emer- 
ton) ( = Araneus corticarius) here designated. 
Bonnet, 1955, Bibliographia Araneonnn, 2: ^17, 



^ Although the starting point of zoological 
nomenclature is Linnaeus' Sijstema Naturae, 10th 
edition, with the arbitraiy date of 1 January 
1758, the work of C. Clerck, 1757, Svenska 
Spindlar, pul)]ished before, is an exception per- 
mitted by Article 26 of the old International Code 
on Zoological Nomenclature passed at the XIII 
International Congress of Zoology of 1948. After 
tlie XV International Congress of Zoology in 
London in 1958 adopted a new Code (1961), 
Clerck's Aranei Suecici was placed on the Official 
List of Accepted Works by Direction 104 of the 
International Commission on Zoological Nomen- 
clature [1959, Bull. Zool. Nomencl., 17(3-5): 
89-91]. 



was in error when he said that McCook failed 
to include species in the sul:)genus. 
Euaranea Archer 1951, Amer. Mus. Novitates, 
1487: 34. Type species for new subgenus; 
Aranea cavatica (Keyserling) by original desig- 
nation. 

Tlw structure of Araneus genitalia. The 
terms used for the sclerites of the x^alpus 
are those of Comstoclc (1910), which have 
become widely used. I used them in the 
revisions of Theridiidae and of Argiope 
( 1968 ) and they were also used by Grass- 
hoff (1968). 

No terms are in general use to describe 
the female genitalia; I here follow Grass- 
hoff. The epigynum has a prominent 
scape (Fig. 1). The scape is attached to 
the base. Below and behind the base are 
a pair of basal lamellae which are large and 
extend on each side of the epigynum of 
Araneus marmoreus (Figs. 1-3). The slit- 
like openings are on the venter and lead 
into a funnel which continues posteriorly 
but is partly open on one side as a groove. 
The groove, as can be seen in cleared or 
macerated preparations, runs into a funnel- 
shaped chamber toward the middle of the 
base and then bends toward the outside 
and continues under the lateral sclerites 
( crosshatched in Figs. 1-3). In A. diade- 
matus a median posterior sclerite ( Fig. 36 ) 
covers the grooves, while the funnels are 
more or less open posteriorly in A. ;?7<7r- 
moreus (Fig. 3). 

The palpus of Araneus has a large 
terminal apophysis (term, apoph., Figs. 4- 
6, 8), a sclerite below the subterminal 
apophysis (subterm. apoph., Figs. 4-6, 8), 
and distal hematodocha ( dist. hemat. ) that 
are absent in Argiope and Gea ( Le\'i, 1968) . 

Species differences. Archer (1951) de- 
scribed and pictured only the median 
apophysis of the palpus (med. apoph. in 
Figs. 4-6, 8) and the scape of the epigy- 
num. He erected numerous araneid genera 
on the basis of the shap)e of the median 
apophysis. Unfortunately, the median 
apophysis turns out to be unreliable for 
separating species of Araneus and is pre- 



134 Bulletin Mmeiim of Comparative Zoology, Vol. 141, No. 4 



sumably a poor character for separating 
genera. Related species often have a 
similar median apophysis (see A. gemma 
group. Figs. 199, 211, or A. saevus and A. 
diadcmatus. Figs. 38, 55). There are also 
indix'idiial differences in this strvicture 
within species (e. g., A. nordmanni. Figs. 
61, 63, 65, 67). This unreliability of the 
median apophysis as a species specific char- 
acter is of interest in view of its definite 
function in copulation: its median spine 
takes hold of the tip of the scape of the 
epigynum (van Helsdingen, 1965; Grass- 
hoff, 196S). The form of the scape also 
varies among individuals of a species. 

The distance between the spine and the 
distal spine or ridge on the median apophy- 
sis varies in different species with the 
length of the epigynal scape. The spine is 
close to the distal end of the median 
apophysis in species having a short scape 
(A. gemma, A. gemmoides, A. pima). The 
distance is large in those having a long 
scape (A. bicentenarius, A. andrewsi, A. 
diadcmatus; and A. saevus). In the related 
genus Eriopliora, in which the particularly 
long scape reaches to the spinnerets, the 
scape is matched by a shift in position and 
elongation of the median apophysis. 

The shape of the conductor is of diag- 
nostic value only in some species: it is an 
unusually large sclerite in A. corticarius 
(Fig. 120), and a very small, narrow one 
in A. trifolium (Fig. 179). The conductor 
in copulation embraces the scape of the 
epigynum, facing the median apophysis, 
forming a track on which the scape can 
glide in two directions. 

Of far greater diagnostic value in sepa- 
rating species arc the shapes of the embolus 
and terminal apophysis (term, apoph. in 
Figs. 4-6, 8), structures easily seen but 
often slighted in illustrations of araneid 
species. (Seen in all illustrations of the 
mesal view of the palpus in this paper.) 
Part of the embolus enters the female duct 
system (the left embolus, the left opening 
and ducts). But during copulation the 
terminal apophysis comes to lie below the 



base of the scape, against the abdomen, 
and supplies support. 

The cap that is present on the embolus 
of virgin males (Levi, in press) differs in 
shape in different species. The caps of 
some species (e.g., A. ilknidatiis) are very 
large and noticeable. 

The paracymbium differs greatly among 
species within the genera Meta and Zijgi- 
eUa (Gertsch, 1964), but is similar in 
almost all Araneus species. This only illus- 
trates the difficulty of making generali- 
zations or of attempting to use only one 
sclerite of the palpus, such as the median 
apophysis, for species diagnosis in all 
genera of the family. The paracymbium, 
during copulation, lies near the base of the 
median apophysis; it prevents the median 
apophysis from turning and by hooking 
into the tegulum border, acts as a stopping 
wedge between median apophysis and 
tegulum. 

In the female the best diagnostic char- 
acters are those of the epigynum. How- 
ever, the length and wrinkling of the 
prominent scape varies, perhaps due to 
different folding. In most species the scape 
is short; however, in some species it is 
always long (bicentenarius, andrewsi, dia- 
dcmatus, and saevus). The general shape 
varies among species. The scape may 
break off during mating in almost any 
species, and usually is torn off in A. corti- 
carius and A. groenlamlicohis. 

The general structure of the base of the 
epigynum is much less variable and differs 
more between species than among individ- 
uals of a species. As it is undesirable to 
cut off the scape to view the base, it is best 
examined in posterior view by pulling the 
epigynum slightly away from the abdomen. 
Unfortunately, most authors have not illus- 
trated this posterior view, exceptions 
being Wiehle (1963) and recently Gertsch 
(1964) in revising American species of the 
araneid genus Zygiclla. The basal lamellae 
are unfortunately also (|uite variable among 
different individuals of a species, as is 
shown by Grasshoff (1968, p. 46, fig. 41). 



Araneus diadematus group Orb-weavers • Levi 135 



However, genitalia are not the only char- placement" is a useful term proposed by 

acters. Araneus corticarius has highly dis- Brown and Wilson (1956) for the phenom- 

tinctive abdominal shape and coloration enon that morphological characters of 

(Fig. 118). It cannot be confused with any related species having overlapping ranges 

other North American species. The cross- show greater difference in the area of 

shaped arrangement of markings on the overlap than in their allopatric distribution, 

dorsum of A. diadematus (Fig. 37) is diag- Examples have been found whenever taxo- 

nostic in North America, although not in nomic revisions are made and there is 

Europe. abundant material. In my theridiid re- 

The second tibia in males of many spe- visions I found character displacement be- 

cies has strong spines and is often bent, tween Tlieridion montanum Emerton and 

Grasshoff (1968, p. 24, fig. 17) illustrated T. lawrencei Gertsch and Archer (Levi, 

the considerable variation within the four 1957a, p. 72). Specimens of Steatoda hes- 

spccies studied. While the spine pattern pera Ghamberlin and Ivie and Steatoda 

is not a good character for separating spe- horealis (Hentz) are more distinct where 

cies of Araneus (L. D. Carmichael, in the borders of their ranges meet, although 

manuscript), the presence or absence of I failed to point this out in my revision 

spines and modifications of tibia 2 may be (Levi, 1957b). Argiope trifasciata (For- 

used to separate some species. skal) males have a noticeably smaller 

The male of most species has a hook on palpus in the area of overlap with the very 

the distal margin of the first coxa (Fig. similar A. florida Ghamberlin and Ivie, 

183), which fits into a proximal depression which has a larger palpus (Levi, 1968, p. 

on the dorsal surface of his second femur, 335, 337). 

coupling the legs during mating. The coxal The possibility of drawing erroneous 

hooks are small and more posterior in some conclusions by ignoring geographic and 

species (A. gemma) and absent in others individual variation and picking up in- 

(A. cavaticus). A cone or spur is present dividual specimens is best showii in 

on the second coxa (Fig. 183) in some Araneus nordmanni. Araneus nordmanni is 

species. The hook on the first coxa and found in the same habitat, on forest trees, 

the corresponding depression are found in as the dark-colored A. saevus. Where the 

species of many araneid genera and do not ranges of the two species overlap, A. nord- 

seem to reflect close affinity. manni is rarely dark, and shows little vari- 

Any diagnostic feature of any animal ation in size and shape. However, outside 
species is subject to variability within a the range of A. saevus, in the southern part 
certain range. This is true also of spiders, of the range of A. nordmanni, it is often 
Guriously, this is denied by some spider black, it varies in size, and in no two males 
specialists who consider any differences in are the embolus (Figs. 69-75) and median 
structure (individual or geographic) to apophysis exactly the same shape. Araneus 
represent a "species difference." The vari- nordmanni is largest in the southwestern 
ability of many characters of four Araneus part of its range. By looking only at the 
species of Europe has been beautifully median apophysis of the palpus of a few 
illustrated by Grasshoff ( 1968 ) . My obser- specimens and noting differences in size 
vations support Grasshoff 's completely: and color. Archer described A. pseudo- 
while sclerites in the palpus ( Figs. 96-99, melaena from the southwest and A. darling- 
103, 104) and all parts of the epigynum toni from southeast. The differences Archer 
show some variation among individuals observed are there, but by examining larger 
(Figs. 76-92), nevertheless there are gaps samples one can easily see that the differ- 
in the variability between species. ences are within the variation of A. nord- 

Character displacement. "Character dis- manni. Perhaps the differences among 



136 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



populations in the southwest are empha- dii.s (Olivier) and tiny A. poUidus males 

sized by the isolation of the habitats in will court A. dkidematiis. In courtship the 

which this forest species can sur\dve. male plucks web threads in a rhythm char- 

Introp-ession. Of considerable interest acteristic for his species. Females do not 

is the Araneus <^emma group of species, respond to the plucking of a male of the 

including fi\'e species north of Mexico. The wrong species. One male of A. diadcmatus 

eastern A. cavaticus is very distinct, but is gave up only after five hours of courting 

closest to A. gemma of the Pacific coast, a female of A. paUidus. 

There is the widespread A. gemmoides Habits. All North American Araneus of 

from the central states and provinces to the diadematus group have similar life 

the Pacific coast. All these species build histories. There is one generation, which 

large webs on bams, houses, and f)orches matures in summer, mates in summer, and 

although their original habitat may have dies in fall after making a loose fluffy egg 

been chffs and entrances to caves. In sac (Plate 1). Spiderlings leave the egg 

addition there are A. pima in Utah and sac in spring (A. diadematus). 

Arizona, and A. illaudatus from Arizona to The virgin male has the embolus capped 

Texas. The habitats of the last two are (Levi, in press). The function of the cap 

uncertain, although they probably are also is not blown, but it is not believed to 

found on buildings. transmit sperm although a duct is at times 

Araneus gemma McCook was split by visible. Males are known to mate several 

Chamberlin and Ivie into three species: times and to survive mating. It has been 

A. gemma, A. gemmoides, and A. pirus. suggested that mating before sperm in- 

With few specimens on hand the judgment duction, as observed in some spiders, may 

was sound. However, any series shows that remove the cap and permit the embolus to 

A. gemma is quite variable, as is A. pirus; function (Kullmann, van Helsdingen, per- 

furthermore, at times A. gemma is collected sonal communication ) . The cap is diag- 

with A. gemmoides. Some females are nostically different in each species. About 

intermediate and cannot be assigned to half the males in collections are virgin and 

either species (Figs. 215-217). Populations still have the cap, half have mated and 

of A. gemmoides are relatively uniform ha\'e lost it. Virgin males wandering in 

except for coloration in the area of overlap search of a mate may be more likely to be 

with A. gemma. Apparently the two spe- collected. 

cies hybridize and introgression is taking In all collections examined, only one 

place with gene flow into the A. gemma male was marked as having been killed and 

population. It is known that males may tiy eaten by a female A. diadematus. The male 

to mate with females of the wrong species, when examined turned out to be an A. 

Broken-off embolus caps are at times found cavaticus, not the partner of the female, 

on the epigynum of a different species. It However, Grasshoff (1964) reports that 

is especially common to find large A. illau- the female of Araneus paUidus has to bite 

dafus tips in the epigynum of A. pima. into the male's al^domen to permit holding 

Apparently the cross of A. gemmoides X on during copulation. In one case the 

A. gemma is fertile. This would produce a female, feeding on an insect, was prevented 

potentially interesting study. It is not fol- from biting; the male could not hold on 

lowed up here. to the female, slid off, and was wrapped 

It is not uncommon to collect females in silk like prey, 

with the wrong males. This happened to All species make an almost vertical orb 

as careful a naturalist as Emerton. Grass- web \\ith about LS-30 radii, with a retreat 

hoff (1964) reports that males of Araneus above to the side of the web in leaves, 

diadeinatus will court females of A. palli- bark, or lichens and connected with the 



Araneus diadematus group Orb-\veavers • Levi 137 




Plate 1. Egg sac of captive Araneus pimo sp. n. from Arizona 



hub by a signal line (Plate 2). During the 
day the spider rests in the retreat, at night 
usually in the center. The center has an 
irregular mesh. Araneus diadematus is 
more likely than other species to be found 
in the center of the web at daytime. Al- 
though all species mature at the same time, 
their variation in size is reflected in the 
diameter of the orb and the size of the 
prey handled. Also each species seems to 
have its own habitat, some preferring 
forests (A. saeviis, A. nordmanni, and the 
small A. corticarius), some meadows (A. 



trifolium, A. marmoreus), city gardens 
(A. diadematus in North America), build- 
ings or cliffs ( A. cavaticus group ) . Araneus 
hicentenarius, often brightly colored, makes 
its retreat among lichens (Plate 3), the 
color of which it matches. 

Species living in the open in herbaceous 
vegetation tend to have an oval abdomen 
(A. marmoreus, A. trijolium), while those 
living on trees and cliffs have humps (A. 
nordmanni, A. saevus, A. cavaticus). The 
adaptation of this curious correlation is not 
known, but is also believed true of 



138 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 




im^^n 



Plate 2. (Top) Web of Aroneus covaf/cus (Keyserling) powdered with corn starch, West Virginia. (Photo by W. A. Shear. 
(Bottom) Web of A. gemmoides Chamberlin and Ivie with adult female, Wisconsin. 



Araneus diadematus group Orb-weavers • Levi 139 




Plate 3. Araneus bicentenarius (McCook) in retreat among 
lichens. West Virginia. Both spider and the lichens are the 
same shade of green. (From Ektachrome photo by W. A. 
Shear.) 



European species. It has been suggested 
(W. S. Shear, personal communication) 
that only those that have a retreat in curled 
up leaves lack humps. 

Western A. gemmoides and some A. 
gemma have black bars on the venter, 
which are distinct "eyes" in A. pima (Plate 
4). I assume that when disturbed the 
spider zooms down its signal thread to 
appear at the center of the orb web. In 
South America I watched an araneid make 
maximum use of its ventral spots, which 
in the particular individual observed were 
in the shape of a face. 

The habitat observations reported here 
are gleaned from the labels on vials. Only 
mature individuals were used. Most check- 
lists and literature on habits are unreliable 
because the species determinations are in 



doubt and often wrong. The specimens 
usually have not been turned over to a 
museum for safe keeping as voucher 
specimens. In mapping distributions, only 
specimens examined have been used. 

Key to Females 

la. Abdomen with humps or angular anteriorly 
(Figs. 21, 37, 118) 6 

lb. Abdomen oval to spherical, without humps 
or anterior angles (Figs. 144, 177) 2 

2a. Posterior lamellae of epigynum showing on 
each side in ventral view as large curved 
folds; scape overhanging a depression 

bordered on each side (Figs. 1, 107) 

marmoreus 

2b. Posterior lamellae not visible in ventral 
view or, if visible, scape not overhanging 
a depression bordered on each side 3 

3a. Scape of epigynum with more or less 
parallel sides, very rarely broken off 
(Figs. 138, 174) 4 

3b. Scape of epigynum tapered toward tip, 
or often broken off (Figs. 159, 167) 5 

4a. Legs banded; a bordered depression on 
each side of epigynal scape in ventral 
view (Fig. 174); no median keel in pos- 
terior view (Fig. 176) trifolium 

4b. Legs not banded; base of epigynum a 
domed area truncate on the posterior side; 
openings posterior and separated by a 
keeled septum (Figs. 140, 142) iviei 

5a. Scape of epigynum widest at base; on 
each side of scape a shallow depression 
with a narrow rim (Fig. 167) ijukon 

5b. Scape, if present, widest close to its mid- 
dle; scape hiding ventral depression (Fig. 
159) and if scape is broken, rims wide 
( Fig. 162 ) groenlandicohis 

6a. Abdominal humps projecting toward sides; 
usually a transverse line between humps 
anterior of which abdomen is usually dark, 
posterior hght (Fig. 118) corticarius 

6b. Humps dorsal, transverse markings not as 
contrasting 7 

7a. Scape long, at least twice as long as width 
of base — 8 

7b. Scape short 12 

8a. Scape coiled (Fig. 52); southwest 

santarita 

8b. Most of scape straight (Figs. 45, 48) 9 

9a. In posterior view of epigynum there is a 
median hght groove (Figs. 17, 20, 29) .__.10 

9b. No such groove in posterior view — -11 

10a. Venter of groove (toward scape) dark and 
swollen, epigynal lamellae on each side 

large (Fig. 29); Pacific coast andrewsi 

10b. Venter of groo\e not swollen toward 
scape, lamellae small (Figs. 17, 20); 



140 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 




Plate 4. Araneus pima sp. n. showing eye spots on venter of abdomen. 



Canada, eastern United States to South- 
west bicentcnarius 

11a. Spider dark, abdomen very dark, black, 
usually with a white lanceolate cardiac 
mark on abdomen (Fig. 51); forests; 
epigynum as in Figs. 42, 44, 45, 47, 48, 
50 saeviis 

lib. Spider light, abdomen almost always with 
light dorsal marks in shape of a cross 
(Fig. 37); city gardens; epigynum as in 
Figs. 34, 36 diadcmatus 

12a. Epigynal scape with transverse annuli 
separated by grooves ( Figs. 123, 124, 187, 
188) _. ,13 

12b. Epigynal scape lacking transverse annuli 
(Figs. 207, 224, 233) 14 

13a. Scape twisted (Figs. 123, 124); base on 
each side of scape with a diagonal rim 
enclosing the median depression (Fig. 
123); Newfoundland to New Hampshire 
— - — — loashingtoni 

13b. Scape straight (Figs. 187, 188); rims of 
depression hidden by scape (Fig. 187) ____15 

14a. Width of scape more than half width of 
epigynum (Fig. 187); epigynum in pos- 
terior view with median triangular sclerite 
(Fig. 189); Eastern United States and 
Canada, on overhanging cliffs, common 
on buildings cavaticus 



14b. Width of scape less than half base of 
epigynum (Figs. 87, 90); epigynum in 
posterior view with heart-shaped depres- 
sion (Figs. 89, 92); widespread in forests 
nordmanni 

15a. Anterior end of epigynal scape constricted 

(Fig. 233); Texas and southwest . illaudatus 

15b. Anterior end of scape is wide, scape tap- 
ered toward tip 16 

16a. Epigynum and scape very small, scape 
triangular (Fig. 195); widespread in cen- 
tral Canada and United States to Pacific 
coast gemmoides 

16b. Epigynum and scape otherwise (Figs. 203, 
205, 218); Pacific coast and Southwest , ..17 

17a. Scape without lateral rims (Figs. 218, 
221); a deep transverse groove in posterior 
view (Figs. 220, 223); California to 
Texas pima 

17b. Scape witli lateral rims, very variable 
(Figs. 203, 205); transverse groove very 
narrow if present (Figs. 204, 206); 
Pacific coast gemma 

Key to Males 

la. First coxa with a hook on distal margin 

facing second coxa (Fig. 183) 2 

lb. First coxa without hook on distal margin ._ 9 
2a. Conductor of palpus with teeth around 
margin (Figs. 23, 24, 31); median 



Aranevs diadematvs group Orb-weavers • Levi 141 







o 


"D 






Q- 


"G 


If 




n 


c 
o 

Q- 


^2 






X 




o 


LU 


**- 3 


!=! F 








■D 



O 



c 
(1) 

> 






O 



— ' 




E 


^ 






D 


D 


o 




CN 














O 


<u 




"o 


> 


-o 


o 




^ 






Q- 




c 


LO 


c 






(U 




(U 






> 


"5 


o 


' 






to 










dl 


c 


o 




E 


s 


o 


<u 












D 






o 






^ 


D 






CD 










D 












E 


-D 


o 




'-5 











o 


-a 










c 






c 


en 


o 


• 




(U 


o 


Q. 


o 




O 


-Q 


X 


^ 




Q. 




LU 


a) 




c 

D 




o6 


> 














■*- 


D 


"D 


CN 




D 


C 








t/T 


o 


c 


E 




3 


^ 


D 


3 




OJ 




Q. 


C 






D 


X 


>. 




o 




UJ 


D) 




E 


W) 




'q. 




o 




1^ 


LU 




E 


o 


^ 


_■ 




*yi 


E 


3 


. — 




3 


*- 


> 


1 




a> 





a> 


o- 




c 
D 


E 


O 

Co 
















< 


<i 


<c 


(1) 




. , 


,.,_ 


■ , 


O 




O 


o 


o 


3 




E 






LU 




3 


3 


■ ■ 




3 


Q. 


_Q. 


-i£ 




C 








O 


O 


U 




a 


O. 


D. 


dj 














'q. 


.-^ 


•^ 


U 




<D 


J) 


<1> 




E 










'o 


'o 


"o 




£ 


a) 


<u 


OJ 


3 


LO 


3 


3 


D 


c 


o 


U 


"u 


*J 


o 


t/> 


D 


3 


D 






(^ 


lyi 


C^ 


-=t 




o 

u 










CO 
1 


1 


CO 




"^ 


rJ, 


°^ > \„ 



O) O) O) 



142 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



apophysis with at least 2 distal spines 
(Figs. 23, 32) 3 

2h. Conductor of palpus smooth around mar- 
gin; median apophysis with a keel on 
distal edge (Figs. 4, 110) 4 

3a. Terminal apophysis with drawn-out tip 
(Figs. 31, 32); teeth around edge of 
conductor suhequal (Fig. 31); Pacific 
coast - andrewsi 

31). Tip of terminal apophysis an acute tri- 
angle (Figs. 23, 24); first tooth on em- 
bolus side of conductor noticeably larger 
than others (Figs. 23, 24); Canada, 
eastern U. S. to Southwest hicentenurius 

4a. Second co.xa with spur (Fig. 183) 6 

4b. Second coxa without spur 5 

5a. Palpal conductor unusually large (Fig. 
120); in mesal view median apophysis 
about 3 times as long as wide (Fig. 119); 
Pennsylvania to Alaska corticarius 

5b. Palpal conductor of normal size (Fig. 
128); median apophysis about 2 times as 
long as wide (Fig. 127); Newfoundland 
to New Hampshire - washingtoni 

6a. Terminal apophysis paralleled b\' a long 
subterminal apophysis (Figs. 4, 96-99, 
103, 104, 110) 7 

6b. Subterminal apophysis a short stub 8 

7a. Embolus sclerotized, subcircular at end 

(Figs. 4, 103, 104, 110); holarctic 

_ marmoreus 

7b. Embolus longer than wide as in Figs. 69- 
75, 96; holarctic nordmanni 

8a. Embolus a curved hook (Figs. 38, 40, 
41); apical apophysis short, curved, sickle- 
shaped (Figs. 38, 39) diadematm 

8b. Embolus otherwise (Figs. 55, 57, 58); 
apical apophysis a long, strongly curved 
prong (Figs. 55, 56) sacvtis 

9a. Inner lamella of embolus a wide projecting 
plate of larger visible area than narrow 
conductor (Fig. 179); embolus with distal 
notch (Figs. 178, 180, 181) trifolium 

91). Inner lamella of embolus, if projecting, 
with visible part always much smaller in 
area than conductor (Fig. 172); embolus 

otherwise 10 

10a. Second tibiae modified by being swollen 

or curved (Fig. 184) 11 

10b. Second tibiae straight, not swollen 12 

11a. Median apophysis with slender, graceful 
proximal spine and distal projections with 
strong, straight spines, the lower one with 

a dentate edge (Figs. 171, 172) yukon 

lib. Median apophysis witli proximal spine 
stout and distal projections both bent out 
at right angles to main axis of median 
apophysis (Figs. 164, 165) .. groenlandicoliis 
12a. Terminal apophysis a sclerotized prong, 
widened just before tip (Figs. 145-148); 



conductor with a pocket on lateral side 
(Figs. 146, 148); embolus as in Figs. 145, 
149-151 iviei 

12b. Terminal apophysis lightly sclerotized, 
fleshy, never widened near tip; conductor 
and embolus of different shape 13 

13a. Males more than 10 mm total length; 
proximal spine of median apophysis placed 
in proximal half of sclerite; distal edge a 
keel (Figs. 191, 192); eastern United 
States, Canada cavatictis 

13b. Males less than 9 mm in total length; 
proximal spine of median apophysis placed 
close to distal spine (Figs. 211, 212, 238, 
239); central, western and southwestern 
region ____ 1 4 

14a. Palpal tibia almost equal in size to bulb 
(Figs. 199, 200); spines of median 
apophysis slender; terminal apophysis 
truncate ( Figs. 199, 200 ) ; widespread in 

central and western North America — 

fienimoides 

14b. Palpal tibia less than half area of bulb; 
spines of median apophysis stout; terminal 
apophysis more pointed; Pacific states, 
southwest to Texas 15 

15a. Embolus pointed (Figs. 211, 214); em- 
bolus cap short (Fig. 213) gciutna 

15b. Embolus truncate ( Fig. 232 ) ; embolus 
cap elongate, pointed (Figs. 231, 240); 
Texas to California - 16 

16a. Total length 5-11 mm; spines of median 

apophysis stout (Figs. 228, 229) pima 

16b. Total length less than 4 mm; spines of 
median apophysis slender, recurved ( Figs. 
238, 239 ) illamlaius 

Araneus angulatus Clerck 
Figures 9—14 

Araneus angulatus Clerck, 1757, Svenska Spindlar, 
p. 22, pi. 1, figs. 1-3, 9. Type specimens in 
the Natural History Museum, Stockliolm, lost. 
Locket and Millidge, 1953, British Spiders, 2: 
127, figs. 80, 83, $, $. Bonnet, 1955, Biblio- 
graphia Araneorum, 2: 433 (in part). 

Aranea angulata, — Wiehle, 1931, in Dahl, Die 
Tierwelt Deutschlands, 23: 52, figs. 4, 5, $, $. 
Roewer, 1942, Katalog dcr Araneae, 1: 795 (in 
part ) . 

Note. Although there are many Hterature 
records of this species m North America, 
all refer to large specimens of various other 
species (A. anclrew.si; A. hicentenarius; A. 
saemis; A. marmoreus; A. nordmanni) that 
had been misidentified. No specimens of 
Araneus angulatus coming from North 
America have been found in collections. 



Araneus diadematus group Orb-weavers • Levi 143 




• H 

O Araneus andrewsi 

• Araneus bicentenorius \, 



Map 1. Distribution of Araneus bicenfenorius (McCook) and A. andrewsi (Archer) 



The differences between A. angulatus and 
Noith American species are discussed un- 
der A. bicentenarius. 

The web, described by Wiehle (1931) 
has bridge threads up to 5 m long at 3 ni 
height. The number of spokes averages 21 
and the snare region is 40-63 cm across. 

Araneus bicentenarius (McCook) 
Plate 3; Figures 15-26; Map 1 

Epeira gigas Leach, 1815, Zoological Miscellany, 
2: 132, pi. 109. Female type probably from 
America in tlie British Museum, Natural History, 
examined. ( Not A. gigas, — Comstock, 1912, 
1940, The Spider Book and some other 
authors^. ) 

Epeira bicentennaria McCook, 1888, Proc. Acad. 
Natur. Sci., Philadelphia, p. 195, figs. 3, 5, $. 
Syntypes from "northwestern Ohio and Al- 
legheny mountains" lost. 

Epeira angulata var. bicentenaria, — McCook, 
1893, American Spiders, 3: 186, pi. 10, figs. 
3-5, pi. 11, figs. 2-4, 5, S. 

Aranea bicentenaria, — Archer, 1951, Amer. Mus. 
Novitates, 1487: 31, figs. 68, 78, ?, $. 

Aranea kisatchia Archer, 1951, Amer. Mus. 



^ The International Commission on Zoological 
Nomenclature will be asked to place the name 
Epeira gigas Leach on the Official List of Re- 
jected Names in Zoology and the name Epeira 
bicentenaria McCook on the Official List of Spe- 
cific Names in Zoology. 



Novitates, 1487: 27, fig. 69, 9. Female holo- 
type from Grant Parish, Louisiana, in the 
American Museum of Natural History, ex- 
amined. NEW SYNONYMY. 

'Note. Leach's description suggested that 
his Epeira gigas is A. hicentenarius. The 
type of E. gigas was found pinned and 
stuffed with cotton in the British Museum 
by Mr. D. J. Clark. After placing it in 
alcohol Mr. D. J. Clark examined the speci- 
men and reported it to be an unusually 
large specimen of A. angulatus; upon com- 
paring it to my drawings he found it to 
match Figures 18, 19 of a southern A. 
bicentenarius. I have since examined it 
myself. The name A. gigas has been used 
by some authors for A. marmoreiis, never 
before for A. Iricentenarius. A junior 
homonym, Epeira gigas C. L. Koch, 1830, 
has been renamed A. grossus (C. L. Koch) 
and belongs to a species also closely related 
to A. angulatus. The spider was collected 
by McCook in 1882 at the bicentennial of 
the city of Philadelphia. By misprint the 
name was spelled with two n's in the first 
publication. 

Description. Female from New Jersey. 
Carapace dark brown. Sternum dark brown 
with a lighter, branched mid-longitudinal 
band. Legs mottled brown with darker 



144 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



bands. Abdomen very dark with folimn. 
Venter dark brown to black, sometimes 
with a median hght area. Abdomen with 
large hnmps. The dorsum has small 
sclerotized spots, the bases of setae, but 
all setae are short. Total length, 15 mm. 
Carapace, 5.9 mm long, 5.0 mm wide. First 
femur, 6.S mm; patella and tibia, 9.2 mm; 
metatarsus, 5.4 mm; tarsus, 1.8 mm. Second 
patella and tibia, 8.4 mm; third, 5.0 mm; 
fourth, 7.6 mm. 

Male from New Hampshire. Coloration 
as in female except for an anterior, dia- 
mond-shaped, longitudinal white mark on 
dorsum of abdomen. The first coxa has a 
hook on the distal margin, the second a 
spur. The second tibia is very strong with 
strong macrosetae. Total length, 7 mm. 
Carapace, 6.5 mm long, 4.6 mm wide. First 
femur, 6.9 mm; patella and tibia, 8.4 mm; 
metatarsus, 4.9 mm; tarsus, 1.7 mm. Second 
patella and tibia, 8.4 mm; third, 5.0 mm; 
fourth, 6.7 mm. 

Variation. Individuals of this rare species 
differ in size and coloration. The largest 
specimens come from the southern United 
States. Females are up to 28 mm in total 
length; one measuring 21 mm in total 
length had the carapace 10.0 mm long, 
8.6 mm wide. The smallest specimen comes 
from Canada; the smallest female measured 
13 mm in total length, carapace 6.1 mm 
long, 5.6 mm wide. Many northern and 
southern specimens are strikingly colored 
on the abdomen (Plate 3, Fig. 22), others 
are just shades of brown in alcohol (Fig. 
21 ) . The scape of the epigynum is variable 
in length, and the median depressed area 
in posterior view is of variable width, 
narrowest in some southern specimens 
(Fig. 20) called A. kisatchiiis by Archer. 
However, intermediate epigyna are com- 
mon. 

Diagnosis. Females of A. bicentenarius 
and A. andreicsi differ from A. diadematus 
and A. saevus, both of which also have a 
long epigynal scape, by having the median 
area of the epigynum in posterior view 
depressed and hght (Figs. 17, 20, 29). 



Males of A. bicentenarius and A. andrewsi 
differ from males of other groups by hav- 
ing the margin of the conductor toothed 
(Figs. 23, 24, 31, and easily seen in apical 
view) and by the shape of the embolus, 
the tip of which can only be seen in ventral 
view (Figs. 26, 33) after removal of the 
conductor. 

The related Eurasian A. angulatus has 
the median, posterior area of the epigynum 
swollen and sclerotized, rather than de- 
pressed (Fig. 11), and the embolus and 
terminal apophysis of the palpus (Figs. 
12-14) differ in shape from those of the 
two North American species. 

In females of the related Pacific coast 
A. andrewsi, the median posterior de- 
pressed area of the epigynum is distally 
(ventrally) black and swollen (Fig. 29); 
that of A. bicentenarius is not swollen and 
is light in color (Figs. 17, 20). Males of 
A. andrewsi have the tip of the terminal 
apophysis sclerotized and pointed (Figs. 
31, 32), while that of A. bicentenarius is 
an acute fleshy lobe (Figs. 23, 24). The 
embolus can be seen under the terminal 
apophysis inside the palpus. The embolus 
differs slightly in shape in different speci- 
mens; its opening is ventral and is partly 
hidden b)' the conductor. 

Note. In collections and literature A. 
bicentenarius has often been called A. 
angulatus, as have large specimens of var- 
ious North American Araneus species. 

Habits. Araneus bicentenarius is found 
on trees in woods. One brightly marked 
specimen was collected by a botanist 
among lichens on jack pine {Pinus banksi- 
ana). The dorsal pattern makes the spider 
disappear among Evernia mesomorpha, 
Parmelia aundenta, and P. caperata. (L. L. 
Darrow, personal communication.) An- 
other was green when collected, and 
"blended perfectly with lichens, etc." on 
a maple tree in West Virginia. (Plate 3) 
(W. A. Shear, personal communication). 
This last specimen was brown in alcohol 
and did not have contrasting markings. 
Most collections of this rare species consist 



Araneus diadematus group Orb-weavers • Levi 145 




o 

O 



> 



^ (1) ^ 



> 



oo O 

— J2 

. E 

U-) LU 
NO 

o 
E ^• 

O o 

^? 

< "2 



Si 

4) CN 



lA « 



E CN 



E 
o 

D 

o 

E 



o 

a- 



en 


o 


> 




n 


(U 






LU 


:^ 


CN 




o 

CN 
1 


CO 
CN 


E 

3 




U"> 




c 








>^ 






Q. 


D5 








Q. 
LU 




o 


O 
Q. 


E 
F 


o 
u 


0) 


CN 
CN 


to 




—1 


1 




> 




CN 


C 








(U 


tn 


CN 




F 




<n 


-^ 


o 





CN 


<u 


■D 




.r 


JO 






u 


O 










C 


4) 


< 


Q. 


o 


t- 




(U 




O 


■^ 


u 


-Q 


J3 




<u 


i/> 








3 




-D 


E 
F 




O 


c: 


ir 




o 


n 


D 








E 




m 



J. 


E 


ri 


> 


r— 




CN 


^ 










*A 




vt 


D 


0) 


Q 


V 


o 


3 




3 


.n 


OO 


CN 


cn 


E 


U- 


CN 


u_ 


UJ 



146 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 




• Araneus saevus 
O Araneus santarita 



Ma 



p. 2. Distribution of Araneus diadematus Clerck, Araneus saevus (L. Koch) and Araneus santarita (Archer) 



of single specimens, very few of which are 
males. 

Distribution. Eastern North America 
from Nova Scotia, Minnesota to northern 
Florida and Texas (Map 1); one male 
came from Coyote, New Mexico. 

Araneus andrewsi (Archer) 
Figures 27-33; Map 1 

Aranea andrewsi Archer, 1951, Amer. Mus. Novi- 
tates, 1487: 31, figs. 63, 64, 82, $, $. Male 
holotype from Claremont, California, in the 
American Museum of Natural History. 



Description. Female. Carapace brown, 
darker on sides. Stcrnnm mottled brown. 
Legs mottled dark brown, indistinctly 
banded. Dorsum of abdomen brownish 
black, mottled, with a folium. Ventrally, 
area between epigynum and spinnerets 
dark gray. Carapace hairy. Abdomen with 
two large humps, very hairy, and with 
many very small sclcrotized plates, the 
bases of setae. Total length, 14 mm. 
Carapace, 6.7 mm long, 5.9 mm wide. 
First femur, 6.3 mm; patella and tibia, 9.2 



Araneus diadematus group Orb-weavers • Levi 147 



mm; metatarsus, 5.1 mm; tarsus, 1.8 mm. 
Second patella and tibia, 8.4 mm; third, 5.1 
mm; fourth, 7.6 mm. 

Male. Coloration like that of female; 
very dark and hairy. Abdomen with dis- 
tinct humps. First coxa with a hook on 
distal margin, second with a large cone. 
Second tibia strong and bent, armed with 
macrosetae. Total length, 11 mm. Cara- 
pace, 5.2 mm long, 4.4 mm wide. First 
femur, 5.9 mm; patella and tibia, 8.4 mm; 
metatarsus, 4.6 mm; tarsus, 1.7 mm. Second 
patella and tibia, 7.6 mm; third, 4.2 mm; 
fourth, 5.9 mm. 

Variation. Many individuals are almost 
black with the folium outline barely visible. 
Females measured 11-22 mm in total 
length, with carapace 5.0-7.8 mm long, 
4.8-6.1 mm wide. Males measured 8-11 
mm in total length; the smallest had the 
carapace 4.7 mm long, 3.6 mm wide. 

Diafi,nosis. This Pacific coast species 
cannot be confused with any others in 
California. It differs from the related 
eastern North American A. hicentenarius 
in details of the genitalia (see under A. 
hicentenarius) . 

Hahits. Specimens have been collected 
in a house basement. Another was in 
curled-up bark of a eucalyptus tree, with 
the web at chest height, strung toward a 
fence below the spider. Other specimens 
were collected on tree bark, and on the 
trunk of a walnut tree {Jiiglans calif ornica) . 

Distribution. Araneus andrewsi is found 
from Oregon to southern California along 
the coast (Map 1). 

Araneus diadematus Clerck, Cross Spider' 
Figures 34-41, 95, 184^186; Map 2 

Araneus diademaius Clerck, 1757, Svenska Spind- 
lar, p. 25, pi. 1, fig. 4. The type specimens in 
the Natural History Museum, Stockhohn, lost. 
Locket and Millidge, 1953, British Spiders, 2: 
127, figs. 84a, 85a, 86a, 87a, 9, i. Bonnet, 
1955, Bibliographia Araneorum, 2: 486. Grass- 
hoff, 1968, Abhandl. Senckenbergischen Natur- 
forsch. Ges., 516: 1-100. 

Epeira diademata, — Wiehle, 1927, Z. Morphol. 



^ Garden Spider in Great Britain. 



Okol. Tiere, 8: 492; Nielsen, 1932, Biology of 
Spiders, Copenhagen, Vol. 2: 421. 
Aranea diadema, — Wiehle, 1931, in Dahl, Die 
Tierwelt Deutschlands, 23: 70, figs. 103-108, 
2 , $ . Roewer, 1942, Katalog der Araneae, 1 : 
797. Kaston, 1948, Bull. Connecticut Geol. 
Natur. Hist. Surv., 70: 249, figs. 779-782, 
9, $. 

Description. Female. Carapace yellow- 
brown, sternum dark brown, coxae light 
brown. Legs yellow-brown banded with 
darker brown. Dorsum of abdomen with 
white marks anteriorly in form of a cross 
(Fig. 37), a folium posteriorly. Venter 
with a median black band and a pair of 
white spots closer to spinnerets than to 
epigastric groove. Dorsum of abdomen 
often with two humps. Female from Mas- 
sachusetts measured total length, 13 mm. 
Carapace, 4.6 mm long, 3.9 mm wide. First 
femur, 5.0 mm; patella and tibia, 6.7 mm; 
metatarsus, 4.5 mm; tarsus, 1.7 mm. Second 
patella and tibia, 5.9 mm; third, 3.5 mm; 
fourth, 5.5 mm. 

Male. Banding of legs less distinct than 
in female and color generally darker. 
Abdomen with a distinct white cross, and 
folium more distinct than in female. The 
abdomen lacks humps. Second tibia modi- 
fied by being stronger and having veiy 
strong short spines. A specimen from 
Massachusetts measured 8 mm in total 
length. Carapace, 3.8 mm long, 3.4 mm 
wide. First femur, 5.2 mm; patella and 
tibia, 7.1 mm; metatarsus, 4.7 mm; tarsus, 
1.5 mm. Second patella and tibia, 5.7 mm; 
third, 3.4 mm; fourth, 5.2 mm. 

Variation. Females vary 6.5-20 mm in 
total length, carapace 3.5-7.2 mm long, 
3.2-6.1 mm wide. Total length of males, 
5.7-13 mm. 

The shape of the abdomen and size of 
humps varies. The epigynal scape may 
vary slightly in length and the posterior 
triangular sclerite (Fig. 36) varies in shape 
and is often much less pointed behind. 
There are also small differences in the 
embolus cun^ature. This species and its 
variation were recently studied by Grass- 
hoff (1968). 



148 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



Diagnosis. Almost all females and many 
males can readily be recognized by the 
white marks arranged in the form of a 
cross on the dorsum of the abdomen (Fig. 
37). No other species in North America 
has the cross mark. In addition, females 
are distinguished by the posterior face of 
the epigynum, which has a median tri- 
angular sclerite with the straight edge 
out, pointed end in (dorsally) (Fig. 36). 
The homologous sclerite in A. saeviis has 
the outer flat edge with a median notch 
(Figs. 44, 47, 50) and the sclerite heart- 
shaped; the notch leads to a depression 
under the scape (Fig. 45) not present in 
A. diadematiis (Fig. 34). Araneus diade- 
matus has the openings usually distinctly 
visible in ventral view anterior of the tri- 
angular plate (Fig. 34). 

Males are separated from all other 
Araneus species by the recurved embolus, 
triangular and pointed at its tip (Figs. 38, 
40, 41), from A. saevus by the short, 
slightly curved terminal apophysis, and by 
the shape of the conductor (Figs. 38, 39). 

Habits. In Europe the species is wide- 
spread and common in woods, gardens, 
and fields. In eastern North America, A. 
diadematus seems to do well only in cities 
in shrubs between houses, a habitat not 
to the liking of any native species of 
Araneus. In the west records are "from 
floor of food store," "web in woodpile," 
and many in cities. It may take two years 
for the animal to mature in Europe (Locket 
and Millidge, 1953). The female is more 
likely to remain in the center of the web 
at daytime than other American species. 
The species has been used for experimental 
research. The web has about 30 spokes 
and is made at up to 1.5 m height. The 
web has been illustrated bv Wiehle, 1927, 
and Nielsen, 1932; the egg sac by Nielsen. 

Distribution. This Eurasian species ap- 
pears introduced, judging by its limited 
distribution in North America and its pre- 
ferred habitat of shrubs surrounding city 
houses. It survives only in a narrow belt 
from Newfoundland south to Rhode Island, 



west to Vancouver and Oregon, more 
abundant on the milder coasts than in the 
Great Plains and mountain states (Map 2). 
Samples of records are St. Johns, New- 
foundland; Quebec; Toronto; Vancouver; 
Victoria; Boston; Newport, Rhode Island; 
Syracuse; Ithaca; Rochester; Detroit; East 
Lansing; Seattle; Portland, and some 
smaller towns. The oldest American col- 
lections are about 100 years old and come 
from the northeast. The species does occur 
in Siberia and Japan. Specimens from the 
USSR, 64 km SSW of Irkutsk on Lake 
Baikal, were examined; they differed more 
from European ones than do North Ameri- 
can specimens, presumably due to geo- 
graphic variation. 

Araneus saevus (L. Koch) 

Figures 7-8, 42-51, 55-60; Map 2 

Epeira saeva L. Koch, 1872, Z. Ferdinandeum 
Tirol Vorarlberg, (3) 17: 323. Male holotype 
specimens from Bad Ratzes, Austria [aliove 
Siiisi, Trentino Alto Adige, Italy], in the British 
Museum, Natiual History, examined. 

Epeira solitaria Emerton, 1884, Trans. Connecti- 
cut Acad. Sci., 9(6): 299, pi. 33, fig. 11, pi. 35, 
fig. 3, $ . Male holotype from Peabody, Mas- 
sachusetts, in the Museum of Comparative Zo- 
ology, examined. 

Epeira silvatica, — Emerton, 1884, Trans. Con- 
necticut Acad. Sci., 9(6): 300 (in part), pi. 35, 
figs. 1-6, ? . Female paralectotype here desig- 
nated, not lectotype. 

Epeira nigra Emerton, 1894, Trans. Connecticut 
Acad. Sci., 14(3): 402, pi. 1, fig. 1, 2, $■ 
Male and female syntypes from Laggan, Can- 
ada ["5000-8500' in Rocky Mountains near 
Canadian Pacific Railway" a station near 
present-day Lake Louise], in the Museum of 
Comparative Zoology, examined. 

Aranea saeva, — Roewer, 1942, Katalog der 
Araneae, 1: 791. 

Aranea solitaria, — Roewer, 1942, Katalog der 
Araneae, 1: 863. Kaston, 1948, Bull. Connecti- 
cut Geol. Natur. Hist. Surv., 70: 250, figs. 
785-786, 796-797, 9, $. 

Araneus saevus, — Tullgien, 1952, Entomol. 
Tidskr., 73: 164, figs. 10, 12, $, i. Bonnet, 
1955, Bibliographia Araneorum, 2: 588. Wiehle, 
1963, Zool. Jahrb. Abt. System., 90: 276, figs. 
84-92, 2, i. 

Description. Female from near Lake 
Louise, Alberta. Carapace, sternum and 



ArANEUS DIADEMATUS GROUP OrB-%VEA\'ERS • LcVt 149 







o 



E 
o 

O 






> 



I 



U 



P 

^ . 
g-E 



o 



O 









^ 
^ 






CN 



o 

>- 

o 

z 



o 

-a 
o 



o 
u 



I 



o 
c 
>^ 

en 



Q. 
o 



u 

o 



3 

o 



.— <u 



o 

0) 
O) □ 

u: E 



D , 

-■ E 
^ £ 



E 
o 



I 
CN 






< ^ 

— O 



E 
E 
un 
o 

o 
B 

u 

c 



LZ i2. 



150 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



legs dark bro\\'n to black with some silver}^ 
hair. Abdonien dark brown with a darker 
folium, usually witli a wlu'te anterior longi- 
tudinal cardiac maik and two ventral white 
spots side by side. Abdomen covered with 
long and short hair. Total length, 13 mm. 
Carapace, 5.4 mm long, 4.6 mm wide. First 
femur, 5.3 mm; patella and tibia, 7.1 mm; 
metatarsus, 4.2 mm; tarsus, 1.7 mm. Second 
pattella and tibia, 6.4 mm; third, 4.2 mm; 
fourth, 6.5 mm. 

Male from same locality. Male a little 
darker than female. First coxa with a hook 
on distal margin, the second with a spur. 
Second tibia swollen and with macrosetae. 
Total length, 9 mm. Carapace, 5.0 mm 
long, 4.0 mm wide. First femur, 5.0 mm; 
patella and tibia, 6.3 mm; metatarsus, 3.5 
mm; tarsus, 1.4 mm. Second patella and 
tibia, 6.1 mm; third, 3.8 mm; fourth, 5.4 
mm. 

Variation. Although the abdomen is 
usually black with an anterior white 
cardiac mark (Fig. 51), at times the color 
is brown, but the white mark may persist. 
Females \'aried in total length 11-17 mm, 
carapace, 4.0-6.0 mm long, 3.2-5.2 mm 
wide; males were 9-11 mm in total length, 
carapace, 5.0-5.2 mm long, 4.0-4.5 mm 
wide. 

D/flgno.s/.s'. Although many specimens in 
collections were misidentified, females can 
be confused only with A. cUadematu.s. The 
abdomen of A. saevus does not have the 
spots arranged in the cross typical of A. 
diadematus and the epigynum differs in 
details (see A. diadematus). 

The male may be confused with A. 
nordmanni, but vmlike nordmanni the 
terminal apophysis, a black prong, is very 
strongly curved (Figs. 7, 8, 55, 56). 

Habits. Araneus saevus is found in 
forests on trees; it has been found on lodge- 
pole pine ( Pinus contorta ) in the Canadian 
Rockies, and on a jooplar tree {PopuJus 
sp. ) in Maine. 

Distribution. Comparison of specimens 
confirmed Wiehle's rcccmt observation 
(Wiehle, 1963) that A. .solitarius of North 



America is the same as the Eurasian A. 
saevus. The species is found in America 
probably from southern Alaska south to 
New York state and Oregon (Map 2). 

Araneus santarita (Archer) 

Figures 52-54; Map 2 

Aranca santarita Archer, 1951, Amer. Miis. 
Novitates, 1587: 24, fig. 65, $. Female holo- 
type from Santa Rita Mtns., Arizona, in the 
American Museum of Natural History, ex- 
amined. 

Description. Female holotype. Head 
area daik brown; posterior and sides of 
carapace lighter. Sternum brown with 
white median longitudinal band. Legs 
bro\\n with bands barely showing. The 
damaged abdomen is slightly reddish in 
color, and has small humps. Total length, 
17 mm. Carapace, 6.0 mm long, 5.9 mm 
wide. First femur, 8.6 mm; patella and 
tibia, 10.0 mm; metatarsus, 6.7 mm; tarsus, 
2.4 mm. Second patella and tibia, 9.6 mm; 
third, 6.0 mm; fourth, 8.8 mm. 

Diagnosis. This may be an atypical 
specimen of A. saevus. It differs by its 
more twisted epigynal scape ( Fig. 52 ) and 
details in j^ostcrior aspect of the epigvnum 
(Fig. 54)." 

DistriJnition. This species is known only 
from the type specimen from Arizona 
(Map 2). 

Araneus nordmanni (Thorell) 
Figures 61-94, 96-99; Map 3 

Epcira nordmanni Thorell, 1870, Remarks on 
.Synonyms of European Spiders, p. 4. ITolotxpe 
from Uppland, Sweden [area north of Stockholm 
inchiding Uppsala]. The type was destroyed 
when the Swedish freighter "Oklahoma" carry- 
ing it in the mail sank off Great Britain in 
December 195.3. Thorell, 187.3, Remarks on 
Synonyms of European Spiders, p. 544. Emer- 
ton, 1885, Trans. Connecticut Acad. Sci., 6: 
301, pi. 33, fig. 6, $; Emerton, 1894, Trans. 
Connecticut Acad. Sci., 9: 403, pi. 1, fig. 2, 
9, $. Nielsen, 1932, The Biology of Spiders, 
Copenhagen, Vol. 1, fig. 38. 

Epeira sihatica Emerton, 1884, Trans. Connecticut 
Acad. Sci., 6: 300, pi. 33, fig. 13, pi. 35, figs. 
1,4, S ■ Male lectotype from Beverly, Massa- 
chusetts, in the Museum of Comparative Zool- 
ogy, here designated, examined. 



ArANEUS DIADEMATUS GROUP OrB-\VEAVERS • Lcvi 151 




Q. 
D 



vO 

NO 



CL 


D 




a 


<D 






s 




c 






0) 


I\ 




<u 


O 


Q. 


^t~ 


* 


O 




to 


u 


Q 


NO 


-C 








o 

•-0 


NO 


^ 


D 


^ 


CO 


u 










CN 





a 

c 

D 


NO 


.^ 








C 




^ 


o 


aJ 




— 


3 


CN 




-C 


m 


o6 


o 




NO 


D 






*o 


00 




v> 


u^ 


l-v 


O 




NO 


S 


Q. 






D 


"a 


uS 


u 


c 


^N. 


j: 


o 






_N 










^ 


< 


o 

c 
o 




NO 
NO 


c 

o 






S 




un 




3 


NO 




"5 

XI 




N^" 


• 


I\ 


E 

UJ 


OJ 




o 


3 




vO 


-C 




1 




o 



s s 



c 


N» 


lU 




> 


NO 

CO 


Z 




NO 


NO 


CN 




Ln 


"o^ 


rNv 




"5 


O 

o 


c 
o 

N 






O 


• — 






U 


< 






CN 

NO 


R 




3 


^ 






D. 


, 






D 


NO 


"o" 




Q. 




-D 




— 


3 


O 


E 


<U 


Q. 


O 


E 










O 


O 




_; 


Q. 


U 


u^ 


_c 




■ — ' 


o 


u 


^ 




d 


O 




o 




^ 


"aj 


I\ 


Q. 


d. 


o 

-C 

[3_ 


CN 

NO 


3 


3 


._ 




o 


> 

D 


c 
c 



3 

o 


-Q 

E 




E 


JD 




3 

c 


■o 

o 

c 


E 

UJ 


e" 

E 


D 
< 


<. 


1 

On 


d 






NO 


t/i 


o 


uS 




o 


NO 


t\ 






1 


1 


n 


D 



LJ_ -^ . — . 



152 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 




Araneus nordmanni 



Map 3. Distribution of Araneus nordmanni (Thorel 



Aranea nordmanni, — Wlehle, 1931, in Dahl, Tier- 
welt Deutschlands, 23: 58, figs. 84, 85, $. 
Roewcr, 1942, Katalog cler Araneae, 1: 801. 
Kaston, 1948, Bull. Connecticut Geol. Natur. 
Hist. Surv., 70: 250, figs. 783-784, 793-795, 
9, 6. 

Aranea darlingtoni Archer, 1951, Amer. Mus. 
Novitates, 1487: 25, figs. 71, 75, 5, $. Female 
holotype from Durbin, West Virginia, in the 
American Museum of Natural History; para- 
types examined. NEW SYNONYMY. 

Aranea pseudomelaena Archer, 1951, Amer. Mus. 
Novitates, 1487: 26, figs. 70, 79, 5, $. Male 
holotype from Long's Peak, Colorado, in the 
American Museum of Natural History, ex- 
amined. NEW SYNONYMY. 

Araneus nordmanni, — Bonnet, 1955, Bibliographia 
Araneorum, 2: 553. Wiehle, 1963, Zoo). Jalirl). 
Abt. System., 90: 271, figs. 74-83, 9,6- 



Note. Archer named American speci- 
mens A. darlinii,toni because he thought 
that European ones lack the basal spur on 
coxa II. Wiehle (1963) illustrated the 
spur, and all European males examined 
had it. Also Archer indicates that "on the 
endal side [of the median apophysis] below 
the spur is a rounded angle, a feature en- 
tirely missing in A. norchiumni." As dis- 
cussed in the introduction, the median 
apophysis by itself is not a good character 
to separate species. 

Araneus pseudomelaena Archer was 
thought to be a valid entity, but extremely 
variable. As more collections were ex- 
amined it was found that those specimens 



Araneus diadematus group Orb-weavers • Levi 153 




ci^ 



fV 


o 


oo 


c 


1 


o 


oo 


5 


• 


ro 


D 


Os 


SI 




F 


c 


zt 


(U 


o 


E 


u 





, 


n 


.■^ 


o 






m 


(U 








o 


^ 


i- 


a> 


(i> 


1 


Ll_ 


t/> 




oo 


-^ 




Ov 






c 


CO 


O 


cs 


TO 




Q) 








n 


o 








0> 








</l 






1 




<N 




00 


CN 



a^ 


00 






rv. 


N, 






"o 


"B 






c 


u 






o 

N 

























< 










0^ 






od 








1 


oo" 






NO 


00 






t\ 


^ 








•0 


X 






CO 







E 


_ 







D 


CO 


_c 




C 


00 






>^ 









cn 


0" 







'q. 


CO 


-C 




LU 




Q. 






hC 


4) 




C-i 


t\ 


u 




CN 

1 




"0 




VO 


"5 









c 


c 




^ 


0) 


OJ 




— 


> 


> 




<u 









o 





<u 


E 


0. 





E 


[3. 




E 






rv 


to 




00 


OJ 












c 
c 
D 


CN 
00 





c 
<u 

E 


-1 






CN 


(U 




J3 





rv 


U 


c 






D 




>o 


rj 




vt 


t--. 






3 







E 




"5" 


E 




E 


<! 














^ 










(/> 











CN 


u 


'^ 





^ 


-c 







t-v 





CN 


c 


(U 


z 


<U 


3 




3 


ID 

N 


O) 




O) 




CN 




to 



154 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



ascribed to pseuclomeloena were all allo- 
patric with A. nordmanni, and also that 
many specimens had to be arbitrarily 
placed in one or the other. The problem 
of A. pseiidomelaena is discussed in 
greater detail in the introduction and in 
the paragraph on variation below. 

Description. A female specimen from 
Montana had carapace light bro\Mi, ster- 
num dark brown, coxae and femora light 
brown, distal segments of legs banded, 
dorsum of abdomen with a distinct folium 
( Fig. 93 ) , venter with a dark brown band 
between epigynum and spinnerets, on each 
side of which is a white bracket. Dorsum 
of abdomen with two humps. Total length, 
8.5 mm. Carapace, 3.2 mm long, 2.5 mm 
wide. First femur, 3.4 mm; patella and 
tibia, 4.3 mm; metatarsus, 2.S mm; tarsus, 
1.2 mm. Second patella and tibia, 3.8 mm; 
third, 2.4 mm; fourth, 3.5 mm. 

A female from Arizona had the carapace 
rich dark brown, sternum dark bro\vn, and 
legs dark brown. Only edges of labium 
and of endites light. Abdomen brown to 
black. Venter with two white spots side 
by side. Sometimes a white longitudinal 
mark between two humps (Fig. 94) on 
abdomen. Total length, 15 mm. Carapace, 
6.5 mm long, 5.6 mm wide. First femur, 
6.S mm; patella and tibia, 8.5 mm; meta- 
tarsus, 5.5 mm; tarsus, 1.7 mm. Second 
patella and tibia, 8.4 mm; third, 5.4 mm; 
fourth, 8.0 mm. 

Male. A specimen from Montana had 
coloration as in female. Venter with a 
black band and two white brackets, or 
brackets often reduced to four white spots. 
The abdomen is longer than wide and 
much wider in front than behind. The 
humps are small. The first coxa has a 
distal hook, the second a spur. The second 
tibia is bent and bears strong macrosetae. 
Total length, 7.5 mm. Carapace, 4.2 mm 
long, 3.2 wide. First femur, 4.6 mm; patella 
and tibia, 5.8 mm; metatarsus, 3.4 mm; 
tarsus, 1.2 mm. Second patella and tibia, 
4.7 mm; third, 3.0 mm; fourth, 4.6 mm. 



A male from Graham Mountains, Ari- 
zona, had coloration like that of female 
from Arizona. Distal segments of legs 
lighter and banded. Abdomen almost black 
with two humps. First coxa with a hook 
on distal margin, second with a spur. 
Second tibia modified with spines and 
swollen. Total length, 10 mm. Carapace, 

6.4 mm long, 5.5 mm wide. First femur, 

7.5 mm; patella and tibia, 9.2 mm; meta- 
tarsus, 5.7 mm; tarsus, 1.7 mm. Second 
patella and tibia, 8.0 mm; third, 5.0 mm; 
fourth, 7.6 mm. 

Variation. The size and coloration of 
this species are variable. The largest and 
darkest specimens are found in the south- 
western part of the range; the size given 
above is for the largest. The total length of 
females is 7-19 mm, the smallest having 
the carapace 3.0 mm long, 2.8 mm wide. 
Males range 6-10 mm, the smallest having 
the carapace 3.1 mm long, 2.8 mm wide. 

Often specimens in alcohol arc strikingly 
colored black and white, the legs con- 
trastingly banded. 

The genitalia of A. nordmanni become 
increasingly variable as one moves south, 
especially the structures of the male palpus, 
including the median apophysis so care- 
fully studied by Archer (Figs.' 61-68). But 
no two males in the southern part of the 
range have the median apophysis or the 
embolus identical (Figs. 69-75, 96-99), 
and as these variable southern popidations 
are allopatric with the less variable north- 
ern A. nordmanni, and many males would 
have to be arbitrarily placed, they are 
considered one species. Except for epigy- 
num size and scape length and shape, the 
epigyna are all alike. 

Diagnosis. In the part of the range in 
which A. nordmanni overlaps with A. 
saevus, the former is smaller than the latter 
species; outside the range of A. .saevus it is 
larger. Females of A. nordmanni have a 
much shorter scape (Figs. 87, 90) than 
those of A. .saevus. The species can be 
confused \\'ith the sympatric A. cavaticus 



Araneus diadem atus group ORB-^VEAVERS • Levi 155 




o 



o 

Q. 



O 

E 



O 

Q. 
D 



D 

E 
-a 



I 



o 

Q. 



J— 


o 




JD 




f- 


•o 


(U 


o 





Q. 
O 
Q. 
D 



<U 

E 
o 

XI 

o 



U 



o 
o 






o 

E 



o 



o 

E 



Q. 
O 

Q- 
O 



O 



_ C 



o 
en 



< i 



o 

Z 



156 Bulletin Museum of Compamtwc Zoology, Vol. 141, No. 4 



in the eastern part of Canada and United 
States; ho\\'ever, in posterior view of A. 
nordmanni the rims of the epigynum en- 
elose a semicircular, heart-shaped depres- 
sion (Figs. 89, 92) that is absent in A. 
covaticus ( Fig. 189). A. norclmanni females 
hav(^ humps on the abdomen, distinguish- 
ing them from females of A. manitobae, 
which have a somewhat similar epigynum. 

The prong of the terminal apophysis of 
the male palp (Figs. 96-99) is less sclero- 
tized and less strongly curved in A. nord- 
manni than in A. soevus. The shape of the 
embolus, which opens on a tip pointing 
away from the cymbium, is quite character- 
istic (Figs. 69-75) and distinguishes all 
male specimens of A. nordmanni from all 
other North American Aranens species. 

Habits. Araneus nordmanni is found on 
trees or, sometimes, boulders, often in dark 
coniferous forests, and often at high ele- 
vations in mountains (for instance, Mt. 
Grizzly, Brit. Col, at 1600-2200 m ele- 
vation and Long's Peak in Colorado). Col- 
lection comments read: On lodgepole pine 
(Pinii.s contorta) in Alberta; on bush in 
Montana; on birch {Betuh sp.) on Mt. 
Katahdin, Maine; on building in Porcupine 
Mts., Michigan; mixed ponderosa pine 
(Pimis ponderosa) and juniper (Jiiniperus 
sp.) in Springerville, Arizona. Nielsen 
( 1932 ) shows a retreat among lichens on 
branches. 

Distribution. Aranens nordmanni was 
first described from Scandinavia. It has 
since only rarely been found in Europe. 
I have examined specimens from Adel- 
boden, Switzerland; Giant Mountains, now 
eastern Poland; Uppland, Sweden; and 
Jarve, Finland. All are much alike and 
much like specimens from the northern 
parts of North America. A new record 
from 64 km SSW of Irkutsk, on Lake 
Baikal, Siberia, USSR, 13. VIII. 1968 (B. 
and J. Hocking), indicates that this species 
does have a wide distribution in Eurasia. 
In North America it seems to occur in 
forested regions north of Mexico (Map 3). 



Araneus marmoreus Clerck, Marbled Spider 
Figures 1-6, 100-105, 107-113, 183; 
Map 4 

Araneus marmoreus Clerck, 1757, Svenska Spind- 
lar, p. 29, pL 1, fig. 2, $ . Female type from 
Sweden in the Natural History Museum, Stock- 
holm, lost. Locket and Millidge, 1953, British 
Spiders, 2: 130, figs. 79d, 84c, 85c, 86c, 87b, 9, 
S- Bonnet, 1955, Bibliographia Araneorum, 2: 
534. Grasshoff, M., 1968, Al)handl. Sencken- 
bergischen Naturforsch. Ges., 516: 1-100. 

Aranea raji Scopoli, 1763, Entomologia Carniolica, 
p. 394. The Scopoli collection was destroyed 
in the 18th century (Horn and Kahle, 1936, 
Entomol Beihefte, 3: 252). Wiehle, 1931, in 
Dahl, Die Tiei"we]t Deutschlands, 23, p. 75, 
figs. 109-114, 2, <5. Roewer, 1942, Katalog der 
Araneae, 1: 802. 

Epeha instilari.s Hentz, 1847, Boston J. Natur. 
Hist. Soc, 5: 470, pi. 30, fig. 10, $. Female 
holotype from an island in the Tennessee River, 
destroyed. Emerton, 1884, Trans. Connecticut 
Acad. Sci., 6: 309, pi. 33, fig. 1, pi. 35, fig. 
18, 9, $. Keyserling, 1892, Die Spinnen 
Amerikas, 4: 170, pi. 8, fig. 126, $. 

Epeira ohesa Hentz, 1847, Boston J. Natiu-. Hist. 
Soc, 5: 471, pi. 30, fig. 11, 9. Female hok)- 
type from Maine, destroyed. 

Epeira marmorea, — Emerton, 1884, Trans. Con- 
necticut Acad. Sci., 6: 307, pi. 33, fig. 2, pi. 
35, fig. 17, 9, $. 

Aranea gigas, — Comstock, 1912, Tlie Spider Book, 
p. 475, figs. 493-498, 9,5; 1940, The Spider 
Book, rev. ed., p. 489, figs. 49.3-498, 9, $. 
Not A. gigas Leach; not A. gigas C. L. Kocli. 

Aranea ttisigia Chamberlin, 1919, Ann. Entomol. 
Soc. Amer., 12: 254, pi. 19, fig. 3, $. Male 
holotype from Chalk Creek, Utah, in tlie 
Museum of Comparative Zoology, examined. 
NEW SYNONYMY. 

Epeira raji, — Kaston, 1948, Bull. Connecticut 
Geol. Natur. Hist. Smv., 70: 257, figs. 81f^822, 
2048-2049. 

Description. Female from Michigan. 
Carapace light brown, sternum dark brown, 
legs light brown, indistinctly banded. Ab- 
domen with dorsum white to yellow or 
orange, having a dark pattern of grays and 
blacks, with black lines around light spots. 
Venter with a black band enclosed by 
white brackets. Total length, 14 mm. 
Carapace, 5.2 mm long, 4.5 mm wide. First 
femur, 5.4 mm; patella and tibia, 6.7 mm; 
metatarsus, 4.0 mm; tarsus, 1.6 mm. Second 



Araneus diadematus group Orb-weavers • Levi 157 




V 

■> 

Araneus marmoreus 







■> 



Map 4. Distribution of Araneus mormoreus Clerck. 

patella and tibia, 6.4 mm; third, 4.0 mm; 
fourth, 6.0 mm. 

Male from Michigan. Coloration like 
that of female with legs more distinctly 
banded. The first coxa has a hook on the 
distal ventral rim, the second coxa a spur. 
The second tibia is swollen and bears 
macrosetae. Total length of a specimen 
from Michigan, 8.4 mm. Carapace, 5.0 
mm long, 3.7 mm wide. First femur, 5.0 
mm; patella and tibia, 6.7 mm; metatarsus, 
3.6 mm; tarsus, 1.2 mm. Second patella 
and tibia, 5.9 mm; third, 3.5 mm; fourth, 
5.0 mm. 

Variation. Total length of females, 9-18 
mm; carapace 2.7-5.2 mm long, 2.3-4.5 mm 
wide. Total length of males, 5.9 mm; 



Carapace 2.9-4.3 mm long, 2.3-3.6 mm 
wide. 

There is considerable variation in color 
and pattern. One female from Texas had 
black bands on its legs. The abdomen has 
a black folium in many females collected 
in Alberta, Montana, Wyoming, Europe, 
and isolated other places (Figs. 101, 102); 
in females from other areas the folium en- 
closes symmetrical light patches (Fig. 100), 
or sometimes it has a reticulated pattern. 
The shape of the terminal apophysis and 
the length of the subtcrminal apophysis 
vary from specimen to specimen ( Figs. 103, 
104). Although the median apophysis of 
A. marmoreus is just as variable as that of 
other Araneus species (Grasshoff, 1968, 



158 Bulletin Museum of Comparotive Zoology, Vol. 141, No. 4 



figs. 36e-h), A. marmoreus has escaped 
being split into several species. 

Diagnosis. The common A. marmoreus 
differs from related A. corticarius and most 
other North American Araneus species by 
having an oval, subspherical abdomen. The 
female can be recognized by the large basal 
lamellae of the epigynum which almost 
always extend on each side beyond the 
base and are visible in ventral view (Figs. 
1-3, 107-109). They also extend at times 
in A. iviei. 

Males, despite the distinct palpus, are 
often misidentified. The terminal apophysis 
is sclerotized and along its proximal side 
is a translucent edge. The terminal 
apophysis is paralleled by a subterminal 
apophysis (Figs. 4, 6, 103, 104, 110). In A. 
nordmanni and A. saevus the subterminal 
apophysis is shorter. The embolus resem- 
bles that of A. corticarius, being semi- 
circular below the tip. The embolus lamella, 
if it shows at all, is in the shape of a flat 
hook (Figs. 4-6, 111), not a piece with 
parallel sides as in A. trifolium (Fig. 179). 
The conductor (Figs. 5, 111) is much 
wider than that of A. trifolium (Fig. 179) 
and has a clearly set off, narrower tongue 
facing the median apophysis. 

Habits. Araneus marmoreus prefers tall 
meadows, and places its webs in grasses or 
low shi"ubs, sometimes on trees. In the 
West it has been collected in light lodge- 
pole pine ( Pinus contorta ) forest, its retreat 
under bark, in a relatively humid place. 
Kaston (1948) reports the egg sac to be 
"13 mm in diameter, a flattened sphere of 
loose white silk. There were 653 orange, 
agglutinated eggs, each about 1.15 mm in 
diameter." The web has been illustrated 
by Comstock (1912, 1940) and Kaston 
( 1948 ) , the egg sac by Comstock. The web 
has 24^29 spokes and is made 50-90 cm 
high in grass. 

Distribution. Araneus marmoreus is hol- 
arctic in distribution. The borders of its 
Eurasian distribution have not been criti- 
cally mapped. The many literature citations 
of misidentified specimens will give a 



mistaken idea of the distribution. In North 
America A. marmoreus is found from 
Alaska to the southern United States ( Map 
4). 

Araneus corticarius (Emerton) 
Figures 1 14-122; Map 5 

Epeiia coiiicaiia Emerton, 1SS4, Trans. Con- 
necticut Acad. Sci., 6: 300, pi. 33, fig. 14, pi. 
35, fig. 9, 9 . Two female syntypes from 
Beverly, Mass., in the Museum of Comparative 
Zoology, examined. McCook, 1893, American 
Spiders, p. 176, pi. 8, fig. 7, $ . Emerton, 1909, 
Trans. Connecticut Acad. Sci., 14: 199, pi. 5, 
fig. 3, S. 

Epeira incestifica Keyserling, 1892, Die Spinnen 
Amerikas, 4: 132, pi. 7, fig. 98. Female holo- 
type from Sitka, Alaska [?] (G. Marx), in the 
U. S. National Museum, examined. 

Aranea corticaria, — Roewer, 1942, Katalog der 
Araneae, 1: 860. Kaston, 1948, Bull. Connecticut 
State Geol. Natur. Hist. Surv., 70: 252, figs. 
800-802, 9, ci. 

Aranea denninp.i Archer, 1951, Amer. Mus. Novi- 
tates, 1487: 30, fig. 81, $. Male holotype from 
The Pas, Manitoba, in tlie American Museum of 
Natural History, examined. NEW SYNONYMY. 

Araneus corticarius, — Bonnet, 1955, Bibliographia 
Araneorum, 2: 470. 

Description. Female from Maine. Cara- 
pace brown with some hairs and some 
irregular marks. Sternum dark brown. 
Coxae light. Legs light, contrastingly 
banded. Anterior of dorsvmi of abdomen 
black with a white cross (Fig. 118). Venter 
black with a white bracket on each side. 
Abdomen with large humps. Eyes large 
and close together. Epigynum with a 
folded scape (Figs. 114, 116). Total length, 
6 mm. Carapace, 2.1 mm long, 1.9 mm 
wide. First femur, 2.5 mm; patella and 
tibia, 3.0 mm; metatarsus, 1.6 mm; tarsus, 
0.8 mm. Second patella and tibia, 2.7 mm; 
third, 1.6 mm; fourth, 2.4 mm. 

Male from Maine. Less contrastingly 
colored than female. Abdomen marked 
like that of female but with less contrast. 
Abdomen has humps laterally. First coxa 
with hook, second without spur. The sec- 
ond tibia is swollen. Total length, 4.7 mm. 
Carapace 2.5 mm long, 1.8 mm wide. First 
femur, 3.0 mm; patella and tibia, 3.5 mm; 



Araneus diadematus group Orb-weavers • Levi 159 




o^ "t; 



<u 




5: 


> 


v> 






3 
Q. 


CM 


, 


□ 


I — 


^ 


Q. 






(U 


3 


"5 


D 


Q. 


Z 


s 


D 


2 




Q. 


o 




(U 
D 


- 


T 

a- 


5 


t/i 


• — 


O 


3 




n 


Q. 


c 


1 


O 


<u 


K 


O- 


E 


CN 


<u 


o 

-D 


^ 


o 


XI 


0) 


E 


O 


^ 


dJ 


E 


1 


"o 


o 

-D 




E 


J3 


H 


u. 


O 






(1) 








1 


CO 


"o 


o 




E 


•- 




0) 

u_ 




o 


^d 


o 




CN 








<u 


if 


'~~ 




Vl 




o 


o 


o 


CN 


tv! 


<U 


O 


' — 


1/1 






o 

a. 


"o 


"5 


U-) 




CN 


<u 


o 


I — 








o 


a 




— 1 


— 1 


"5 


00 


UO 


0) 


o 








1 — 


o 



en 



CK 

O Q. 

o -^ 

, — 3 

o 



U CO 



Q. 

o 



> 



E ■- 



E S 

C 



-I 

^ CN 



o 3 

■^ o 

<U J3 

E E 






. O 

E •£ 



C3) 



CN Q. 

^ o 

CO 



o n ^ 



E ;^ 



icN 

o ^ 



.:i CN 

^ , — 
o 



CN 

6. ^ 



o _B 

D) O 

? E 



E 
E 

in 
o 

CD 



o 
"5 

O 






CN 

o .— 



n 




CN 


(U 


< — 


n 


CN 


> 


1 


F 






K 


UJ 


■^ 


o 


O 









CN 



n o 

CN ^ 

■ — c 



E 
E 

o 

5 

o 



LL. ,— U_ ^ 11. .— _ 



160 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 




Araneus corticari us 



Araneus washing toni 



Map 5. Distribution of Aroneus corticari'us (Emerton) and Aroneus wos/iingfon/ sp. n. 



metatarsus, 1.8 mm; tarsus, 0.(S mm. Second 
patella and tibia, 3.0 mm; third, 1.8 mm; 
fourth, 2.5 mm. 

Variation. Total length of females varied 
between 5.2-8.0 mm, the carapace 2.1-2.9 
mm long and 1.9-2.3 mm wide. Total 
length of males, 4.2-5.2 mm. The shape of 
the abdomen is variable. 

Diagnosis. The shape of the abdomen, 
with the humps projecting toward the 
sides, is not seen in any other large North 
American Araneus including A. washing- 
toni. The epigynum of the female (Figs. 
114-117) resembles that of A. marmoreus, 
but the epigynal lamellae are smaller, only 
rarely visible in ventral view, and the 
median field behind the scape between the 
rims is wider than in A. marmoreus. The 
scape usually breaks off during mating. 

The embolus shape (Figs. 119, 121, 122) 
resembles that of A. marmoreus, but the 
enormous conductor, its length half that 
of the bulb in ventral view (Fig. 120), 
and the longer median apophysis (Figs. 
119, 120) separate the species from A. 
marmoreus and A. ivashingtoni. 



Halrits. Kaston (1948) reports having 
collected A. corticarius in moist meadows 
and woods. It has been collected from a 
cranberry bog in Massachusetts, from a 
swamp in Wisconsin, swept from a hem- 
lock seedling ( Tsuga canadensis ) in a 
swamp in Pennsylvania. 

Distribution. Araneus corticarius is found 
from Alaska to New England. Its known 
range matches the range of black spiiice 
(Picea mariana) and tamarack {Larix lari- 
cina), both bog inhabitants. The southern- 
most records are Ice Mine, Potter Co., 
Pennsylvania, and Beverly Shores, Porter 
Co., Indiana; the northernmost, Rampart 
House, Yukon Terr. (Map 5). 

Araneus Washington} sp. n. 
Figures 123-130; Map 5 

Holotype. Male from 3 miles up road, 
Mt. Washington [no date] (J. H. Emerton), 
in the Museum of Comparative Zoology. 
The species is named after George Wash- 
ington, as is the mountain. 

Description. Female. Carapace brown, 
sternum dark brown. Coxae light brown, 



Araneus diadematus group Orb-weavers • Levi 161 




I 

CO 



CO 



CO 



o 
<u 

6 



>- _ 



o 

E 

O 






r- ID 
C 

E 2 
o 

-a . 

D Ic 

E .y 

0) 5 



5 ° 



<u E 

c ^ 

a o^ 



CO 

:^ J 

o 

_:. -o 

J2 E 

i« • 

U I— 

a t 

S 2: 



2; ■£ 

. > 



<» r— 



< S 






E — ^ 3 

D) -7; ^ O != 

•i-§ ci ■? > 

7 CO t .1^ 

^ •— ^ 00 i/> 

CO 5 ^ - 



c: Q. tv 



? > 

U CO 



lo CO 



o 

. Q. 
CO (D 

CO 



E 
o 

O 

e" 

E 



01 r— 



a. ^ 
o o 

■B i 



o 
Lo in 



,- o ^ 






> 



162 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



legs brown, indistinctly banded. Dorsnin 
of abdomen with a folium and a black line 
between humps, darker anterior of line 
with a white cardiac mark and a trans- 
verse white line (Fig. 126). Venter of 
abdomen black with a white bracket on 
each side. Each side of abdomen has a 
black patch which fades toward the ven- 
ter. The shape of the abdomen resembles 
that of most other species having humps; 
the humps are on the dorsum rather than 
the sides. Total length, 6.5 mm. Carapace, 
3.5 mm long, 2.5 mm wide. First femur, 
3.0 mm; patella and tibia, 4.0 mm; meta- 
tarsus, 2.3 mm; tarsus, 1.0 mm. Second 
patella and tibia, 3.5 mm; third, 2.2 mm; 
fourth, 1.0 mm. 

Male. The coloration is like that of the 
female except that the legs are banded. 
The first coxa has a hook, the second lacks 
a spur. The second tibia is swollen and 
has strong setae. Total length, 5.0 mm. 
Carapace, 2.5 mm long, 2.0 mm wide. First 
femur, 2.9 mm; patella and tibia, 3.5 mm; 
metatarsus, 2.0 mm; tarsus, 0.7 mm. Second 
patella and tibia, 3.0 mm; third, 1.5 mm; 
fourth, 2.2 mm. 

Dkignosis. The female differs from A. 
corticarius in having the humps of the 
abdomen dorsal rather than lateral (Fig. 
126). The anterior section of the scape of 
the epigynum is posteriorly directed in A. 
icashingtoni (Figs. 123, 124); in A. corti- 
carixis it is directed anteriorly and then 
folds back (Figs. 114-116). The male dif- 
fers from A. corticarius by the dorsal 
position of the abdominal humps, by the 
much smaller conductor (Fig. 128), and 
the shorter median apophysis (Figs. 127, 
128). 

Hahits. Emerton collected the speci- 
mens on low bushes. The label was written 
in India ink and at some later time Emer- 
ton pencilled "corticaria" behind the inked 
Epeira. The female from New Brunswick 
was collected on balsam fir {Abies hal- 
samea ) . 

Distribution. Newfoundland to New 
Hampshire (Map 5). 



Records. Newfoundland: Gambo, 25. IV. 
1949, 9 (E. Palmen); Eddies Cove, west 
St. John Bay, 30. VII. 1949, 9 (E. Palmen). 
Neiv Brunswick: Green River Forestry Sta- 
tion, 22 July 1965, 9; 18 July 1967, 2^ (T. 
R. Renault). Neto Hampshire: 29, 2<5 
paratypes collected \\\\\\ the holotype. 

Aroneus alsine (Walckenaer) 
Figures 131-137 

Aranca alsine Walckenaer, 1802, Faune Paiisienne, 
2: 193. Type for the Paris, France, vicinity, 
lost. Wiehle, 1931, in Dahl, Die Tierwelt 
Deiitschlands, 23: 83, figs. 119-123, 9, $. 
Roewer, 1942, Katalog der Araneae, 1: 781. 

Epeira alsine, — Wiehle, 1927, Z. Morphol. Okol. 
Tiere, 8: 493. 

Araneus alsine, — Locket and Millidge, 1953, Brit- 
ish Spiders, 2: 133, figs. 85d, 86d, 9, <5 . 
Bonnet, 1955, Bibliographia Araneonim, 2: 429. 

This Eurasian species, distributed from 
Europe to Kamchatka, has not been found 
in North America. It is very similar to A. 
iviei of North America. Araneus alsine is 
found on grasses of moist woods or clear- 
ings, and the hub of the web is about 15-20 
cm above the ground. The rolled-up leaf 
or leaves which are used as a retreat are 
pulled up from the ground (Wiehle, 1927). 
A web with 20 spokes was illustrated by 
Wiehle (1931). 

Araneus iviei (Archer) 
Figures 138-151; Map 6 

Aranea iviei Archer, 1951, Amer. Mus. Novitates, 
1487: 33, fig. 53, 9. Female holotype from 
Keene Valley, Essex Co., New York, in the 
American Museum of Natural History, ex- 
amined. 

Aranea sachimau Archer, 1951, Amer. Mus. Novi- 
tates, 1487: 33, fig. 55, 9. Female holotype 
from Norwell, Plymouth Co., Massachusetts, 
in the American Museum of Natural History, 
examined. NEW SYNONYMY. 

Description. Female paratype of A. iviei 
from New Jersey. Carapace, sternum, legs 
orange-brown without any marking. Abdo- 
men whitish with an orange cast, and with 
a few whitish spots framed by darker 
orange. Venter between epigynum and 



Araneus diadematus group Orb-weavers • Levi 163 













O Araneus yukon 

• Araneus groen landicola 



Mao 6. Distribution of Araneus iviei (Archer), Araneus yukon sp. n., and Araneus groen/andico/us (Strand) 



spinnerets has a white square with orange 
center. Abdomen lacks humps. Total 
length, 12 mm. Carapace, 5.0 mm long, 
4.2 mm wide. First femur, 4.4 mm; patella 
and tibia, 5.4 mm; metatarsus, 3.2 mm; 
tarsus, 1.5 mm. Second patella and tibia, 
4.9 mm; third, 2.9 mm; fourth, 4.6 mm. 

Male from Michigan. Carapace and 
sternum brown, darker on each side. Legs 



brown, very indistinctly marked. Dorsum 
of abdomen whitish with outline of a 
folium. Brownish venter has a light mark 
followed by a dark mark between epi- 
gastric furrow and spinnerets. Coxa with- 
out hooks or spurs. Second leg not 
modified. Total length, 6.7 mm. Carapace, 
4.2 mm long, 3.2 mm wide. First femur, 
4.9 mm; patella and tibia, 5.2 mm; meta- 



164 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



tarsus, 3.6 mm; tarsus, 1.4 mm. Second 
patella and tibia, 4.2 mm; third, 2.9 mm; 
fourth. 4.1 mm. 

Variation. Females vary from 8.5-12 
mm in total length, carapace 4.1-5.0 mm 
long, 3.2-4.2 mm wide. Males varied from 
.5-7 mm in total length, carapace 3.6-4.2 
mm long, 2.9-3.2 mm wide. In posterior 
view no two epigyna are exactly alike 
(Figs. 140, 142). Figure 142 illustrates a 
specimen designated as paratype of A. 
.sachimau by Archer. 

Diaii,nosis. Araneus iviei superficially 
resembles A. trifoUuni; the genitalia are 
unlike those of any other North American 
species but very similar to those of the 
Eurasian A. alsine. The epigynum, unlike 
that of any other North American species, 
has a strongly domed base anteriorly and 
is truncate posteriorly (Figs. 138, 141). In 
A. iviei the shape of the median septum in 
posterior view narrows dorsally toward the 
abdomen (Figs. 140, 142), while in A. 
alsine it \\'idens into a flat plate more or 
less truncate toward the abdomen (Figs. 
133, 134). 

The male's palpal conductor has a distal 
"pocket" (Figs. 146, 148) unlike other 
Northern American Araneus species. It 
differs from A. alsine in the shape of the 
conductor, and in the shape of the embolus 
(Figs. 145-150). 

Habits. The species has been collected 
from a cedar (Thuja occidentalis) swamp 
in Michigan, sweeping old fields in open 
forest vegetation on Isle Royale, on base 
of lodgepole pine (Pimis contorta) and in 
lodgepole woods in Alberta, and in conifer- 
aspen (Populus tremuloide.s) with pockets 
of black spruce (Picea mariana) with 
dense undercover of rosebushes, grasses, 
and sedges in the damp area, at Lake 
George, Alberta (R. E. Leech, personal 
correspondence ) . 

Distribution. From Alberta southeast to 
Pennsyhania. The northernmost record is 
Riverton. Manitoba, the southernmost Len- 
hartsville, Berks Co., Pennsylvania (Map 
6). 



Araneus quadratus Clerck 
Figures 152-158 

Araneus quadratus Clerck, 1757, Svenska Spind- 
lar, p. 27, pi. 1, fig. 3, $. Female holotype 
from Sweden in the Natural History Museum, 
Stockholm, lost. Locket and Millidge, 1953, 
British Spiders, 2: 130, figs. 84b, 85b, 86b, ?, 
S . Bonnet, 1955, Bibliographia Araneorum, 
2: 575. 

Aranea rcaumurii Scopoli, 1763, Entomologia Car- 
niolica, p. .393. Types from Austria. Scopoli's 
collection was destroyed about 1776. ( Horn 
and Kahle, 1936, Entomol. Beihefte, 3: 252.) 
Wiehle, 1931, in Dahl, Tierwelt Deutschlands, 
23: 79, figs. 115-118, 5, c^ . Roewer, 1942, 
Katalog der Araneae, 1: 804. 

Epeira quadrata, — Wiehle, 1927, Z. Morphol. 
Okol. Tiere, 8: 496. Nielsen, 1932, Biology of 
Spiders, 2: 292. 

Note. This European species closely 
related to A. yukon has not been found in 
America. Japanese specimens alleged to 
be A. (juadratus were examined, but large 
differences in the shape of the conductor 
and median apophysis make me think 
neither specimen is correctly identified. 
Araneus quadratus reported from Green- 
land are A. groenlandicolus (Strand). The 
web is made in high grasses of moist areas, 
the center about 50 cm above the ground, 
and has about 20 spokes (Wiehle, 1931; 
Nielsen, 1932). 

Araneus groenlandicolus (Strand) 
Figures 159-166; Map 6 

Epeira (juadrata, — S0rensen, 1898, Vidensk. Med- 
del. Naturhist. Foren. Kobenhavn, 1898. Not 
A. quadratus Clerck. 

Aranea reauinuri var. groenlandicola Strand, 1906, 
Fauna Arctica, 4: 458. New name for speci- 
mens of Epeira cptadrata described by S0rensen 
from Greenland. Female lectotype here desig- 
nated from Ivigtut, Greenland, 15. VHl. 1889 
(Lundbeck) in the Universitetets Zoologiske 
Museum, Copenhagen, examined. 

Aranea manitohae Archer, 1951, Amer. Mus. 
Novitates, 1487: 37, figs. 51, 59, 62, $. Male 
holotype from The Pas, Manitoba, in the 
American Museum of Natural History, ex- 
amined. NEW SYNONYMY. 

Description. Female. Carapace light 
brown with median longitudinal darker 
band and a dark band near each margin. 
Distal tips of chelicerae brown. Labial 



Araneus diadematus group Orb-weavers • Levi 165 






<u 


CN 


<u 
> 


U- 


r^ 




n 




-o 


•o 




■n 


■— 


o 




^ 


(U 


__■ 


o 


^ 


o 








<u 




<u 




■ — 



tA 








o 






z> 








Q. 






O. 




D 










o 




(U 




<U 






Q. 




D 




o 






<U 








2 






D 




O 










E 




^0 




cri 

i\ 






.-*— 








■ — 






<u 








1 










"5 




R 






00 




C 




' 






Ln 




(U 










"^ 




> 




c 






IT) 








CJ 






UO 




OJ 




E 






' 




•o 




o 






^ 








XI 






o 




CN 




o 










Ut 










(U 








<u 














o 






o 
a. 




c 




2 






^" 




_o 




CD 






LO 




c 




l\ 










<u 

(U 




^ 






"o 




o 




O 


















(U 








■^ 






o 




CN 




V 










N3 




o 






ro 








O- 






U-) 




^^ 










« — 




P 




c>.' 










o 


Q. 


^o 






"5 




<u 


O 












c 


u 






















c 




c 


^ 


"o 






<D 




s 










> 




i 


o 






CN 






(/i 


— 1 






LT) 




CN 


3 

"5 

XI 

E 


od 






E 




LD 


LU 








3 






















"5 






O) 




E 




c 




E 


Q. 




3 




OJ 




E 


LU 




C 


• 


> 




'a- 




CD 


o 


rv! 




IT) 


LO 




'd. 


c 




(/> 


1 — 


d 


LU 


<u 


^ 




C 


1 
CN 


D 
u 




> 


^ 




(1> 


LO 


3 


CN 
N3 


ui 
^o 


E 




o 




o 


1 




D 




XI 


,_^ 


j: 


O- 




C 




D 


<u 




LO 




>^ 






Q. 
O 


^ 




"o 


'q. 




O^ 


3 






0) 


LU 






LU 


00 


^ 


s 






E 






c 




CS 




E 


-i; 




o 


■^' 


^O 






u 


Q. 


u^ 


vO 


i^ 




in 


U 


o 


' 




^0 




o 


u 




vt 












_2 


3 






</t 


2 
o 


^ 


Q. 






a. 


^ 


o 
u 

'-0 


O 
Q. 


C 


Q. 
O 
U 


3 


D 

cr 

3 


tv! 


c 

_o 

c 

0) 

o 


O 

5 


Q- 
</> 

c 
o 


3 

o 


o 

E 


c 


z> 


D) 


N3 
S3 


3 


i 


S 


a 







T 




3 


E 


< 


6 

UJ 


< 




< 


o 

E 

LU 




CO 


od 


^C) 


C 


CO 


irt 




to 


1 


E 


T 


CO 


o 




■ — 


O^ 


t-L 


:\ 


o 


iC 


LT) 


o 

-a 


<j 


' 






lO 




XI 






c 




■ — 


<D 


o 




"5 










OJ 






o 


3 




3 




3 


c 


N 


D) 


o 


Ol 


"o 


C3) 


<u 




ii- 




Li- 


E 


U_ 


> 


LO 



166 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



endites almost black except anterior border. 
Sterauni black with a small, anterior 
median light mark. Legs very strongly 
banded. Abdomen white or red with white 
dorsal spots (Fig. 163). Venter is dark in 
front of spinnerets, light behind epigynum, 
and has two longitndinal bands. Abdomen 
without humps. Total length, 9 mm. Cara- 
pace, 4.0 mm long, 3.3 mm wide. First 
femur, 3.5 mm; patella and tibia, 4.3 mm; 
metatarsus, 2.7 mm; tarsus, 1.1 mm. Second 
patella and tibia, 4.5 mm; third, 2.3 mm; 
fourth, 3.5 mm. 

Male holotype. Carapace, legs light 
brown. Sternum light brown with median 
longitudinal white line. Abdomen all white 
with white pigment spots around spin- 
nerets. Coxae without hooks or spurs. Tibia 
of second leg swollen and with strong 
spines. Total length, 6.1 mm. Carapace, 
3.2 mm long, 2.6 mm wide. First femur, 
3.2 mm; patella and tibia, 4.0 mm; meta- 
tarsus, 2.7 mm. Second patella and tibia, 
3.0 mm; third, 1.9 mm; fourth, 2.9 mm. 

Variation. The scape of the epigynum 
of most females is broken off ( Fig. 162 ) , 
apparently it tears when mating. Females 
varied from 9-12 mm in total length, males 
from 5.5-7.5 mm. 

Diagnosis. Araneus groenhndicolus lacks 
humps on the abdomen. Females have 
been confused with A. trijolium because of 
the reddish coloration of the abdomen and 
the median dark carapace stripe; however, 
the epigynum of A. groenlandicolus is very 
different (Fig. 159-162). The epigynum 
has wider rims than that of A. quadratus; 
the inside edges of the rims are covered 
by the scape (Fig. 159) (often broken off). 
The male of A. groenJandicolus has strongly 
bent distal hooks on its median apophysis 
(Figs. 164, 165) differing from those of 
A. quadratus and A. yidwn and other 
Araneus species. 

Habits. Almost nothing is known about 
the habits of A. groenlandicolus, but it is 
assumed to prefer open ground to forest 
as do other species of Araneus with a 



round abdomen. In Alberta it has been 
found in sedge. 

Distribution. Alberta to Greenland and 
south to Minnesota and Maine (Map 6). 
Specimens examined from Greenland came 
from Ivigtut and Godthaabsfjord. 

Araneus yukon sp. n. 

Figures 167-173; Map 6 

Holotype. Male from Snag, lat 62° 24', 
long 140° 22', Yukon Territory, 24 July 
1948, in the American Museum of Natural 
History. The specific name is a noun in 
apposition. 

Description. Female para type. Cara- 
pace brown with darker median and lateral 
bands. Sternum very dark brownish black 
with median longitudinal line of white 
pigment. Legs brown with very distinct 
dark banding. Abdomen with a folium 
posteriorly, a median light longitudinal 
mark anteriorly. Venter light with two 
dark spots side by side and a dark ring 
around spinnerets. Spinnerets blackish 
brown. Abdomen oval to subspherical 
without humps. Total length, 10 mm. 
Carapace, 4.2 mm long, 3.5 mm wide. First 
femur, 4.0 mm; patella and tibia, 4.9 mm; 
metatarsus, 3.0 mm; tarsus, 1.4 mm. Second 
patella and tibia, 4.4 mm; third, 2.8 inm; 
fourth, 4.0 mm. 

Male holotype. Carapace with indistinct 
markings. Sternum dark with a median 
white line. Leg banding indistinct. Dor- 
sum of abdomen has two pairs of white 
spots anterior to a dark folium and an 
anterior longitudinal white mark (Fig. 
170). Ventral dark spots are fused to each 
other and the posterior one to the dark 
ring around the spinnerets. There are no 
hooks or spurs on coxa. The second tibia 
is slightly swollen. Total length, 6.5 mm. 
Carapace, 3.4 mm long, 2.8 mm wide. First 
femur, 3.5 mm; patella and tibia, 4.4 mm; 
metatarsus, 3.0 mm; tarsus, 1.3 mm. Second 
patella and tibia, 3.4 mm; third, 2.0 mm; 
fourth, 3.2 mm. 

Diagnosis. Araneus yukon palpus differs 



Aranevs diadematus group Orb-weavers • Levi 167 



from that of related A. quodrotus in having 
teeth on the lower distal lobe of the median 
apophysis (Fig. 172) instead of a straight 
carina; the conductor of A. yukon is wider 
and the "upper" lobe of the embolus (left 
in Figs. 171, 173) is wider than in A. 
quadratus (Fig. 158). European specimens 
showed little variation in these characters. 

Habits. Firth River, a locality in which 
the species is found, is on the north slope 
of the British Mountains, 25 miles from the 
Arctic Ocean. Although north of the tree 
line, pockets of spruce ( Fica sp. ) to 3-4 m 
high occur in sheltered spots along the 
creeks; probably the specimen was found 
in these (R. E. Leech, in letter). 

Records. Yukon Territory. Female para- 
type collected with holotype; Firth River, 
British Mountains, 9 paratype, 24. VII. 
1956 (R. E. Leech) in the Canadian 
National Museum. 

Araneus trifolium (Hentz), Shamrock Spider 
Figures 174-182; Map 7 

?Epeira viilpecula Walckenaer, 1841, Histoire 
Naturelle des Insects Apteres, 2: 69. Syntypes 
are Abbot, Georgian Spider Illustrations, figs. 
131, 356 from Georgia in the British Museum 
of Natural History, examined. A nomen dubituu 
and a nomen ohlitum. 

?Epeiia appioximata Blackwall, 1846, Ann. Mag. 
Natur. Hist., 17(1): 80. Specimen from vicinity 
of Toronto, lost. A nomen dtthium and nomen 
oblittmi. 

Epeira trifolium Hentz, 1847, ]. Boston Soc. 
Natur. Hist., 5: 471, pi. 31, fig. 1, 2. Types 
from Maine, destroyed. Emerton, 1884, Trans. 
Connecticut Acad. Sci., 6: 306, pi. 33, fig. 8, 
pi. 35, figs. 13, 14, 21, 22, $, $. McCook, 
1893, American Spiders, 3: 145, pi. 1, figs. 3-6, 
pi. 2, fig. 3, 9, S. Kaston, 1948, Bull. Con- 
necticut Geol. Natur. Hist. Surv., 70: 258, 
figs. 823-825, 2047. 

Epeira aureola Hentz, 1847, J. Boston Soc. Natur. 
Hist., 5: 471, pi. 31, fig. 2, 9. Type from 
Maine, destroyed. 

Epeira trifolium var. candidans McCook, 1893, 
American Spiders, 3: 146, pi. 1, fig. 4, ?. Fe- 
male holotype from California, lost, the color 
of the abdomen was yellowish-white. 

Aranea trifolium, — Comstock, 1912, The Spider 
Book, p. 479, figs. 501-508, $ ; 1940, rev. ed.. 
The Spider Book, p. 493, figs. 501-508, $. 
Roewer, 1942, Katalog der Araneae, 1: 863. 



Aranea gosogana Chamberlin, 1920, ]. Entomol. 
Zool., 12: 8, pi. 4, fig. 6, $. Female holotype 
from desert region in California, in the Museum 
of Comparative Zoology, examined. 

Araneus trifolium, — Bonnet, 1955, Bibliographia 
Araneorum, 2: 614. 

Note. Epeira jospidata Walckenaer, 1837, 
has been placed as a synonym, first by 
McCook: the type. Abbot, fig. Ill, how- 
ever, seems to have humps. Abbot's figure 
is either A. nordmanni or A. bicentenarius. 
However, Walckenaer indicated in the de- 
scription of the figure that the abdomen 
was round. 

Description. Female. Carapace brown 
with a lighter band on each side of a dark 
median longitudinal band; sides dark with 
a lighter brown border. Sternum dark 
brown. Legs brown with strongly marked 
darker bands. Dorsum of abdomen reddish 
with white spots ( Fig. 177 ) , venter usually 
reddish brown without markings in adult. 
A specimen from Ithaca, New York, mea- 
sured 15 mm in total length. Carapace, 6.5 
mm long, 5.5 mm wide. First femur, 6.3 
mm; patella and tibia, 8.0 mm; metatarsus, 
5.5 mm; tarsus, 1.7 mm. Second patella 
and tibia, 6.9 mm; third, 4.2 mm; fourth, 
6.7 mm. 

Male. Carapace, sternum and legs brown, 
abdomen whitish, sometimes all white, 
sometimes with ventral marks. None of the 
coxae have hooks or spurs. The tibia of 
the second leg is only slightly thicker than 
that of other legs. A specimen from Mon- 
tana measured 7.5 mm in total length. 
Carapace, 4.2 mm long, 3.4 mm wide. First 
femur, 4.6 mm; patella and tibia, 5.1 mm; 
metatarsus, 3.5 mm; tarsus, 1.5 mm. Second 
patella and tibia, 4.5 mm; third, 2.7 mm; 
fourth, 3.9 mm. 

Variation. Many specimens have the 
dorsum of the abdomen white. Total length 
of females varied between 9-20 mm, with 
carapace 4.0-6.8 mm long, 3.6-5.4 mm 
wide. Total length of males, 5-8 mm, with 
carapace 3.0-3.6 mm long, 2.5-3.0 mm 
wide. 



168 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 







Araneus trifolium 




Map 7. Distribution of /Araneus trilolium (Hentz). 

Diagnosis. Araneus trifolium is not close 
to any other species, although females have 
been confused with A. (iroenlandicolus and 
A. iviei on account of the spherical abdo- 
men, red coloration, and dark leg bands. But 
the epigynum of A. trifolium is very dis- 
tinct: a scape (only rarely broken off) 
flanked on each side by a depression in 
the base; the margin of the depression and 
the depression itself are sclerotized and 
black posteriorly (Fig. 174). 

The male palpus of A. trifolium has the 
embolus of an unusual shape (Figs. 178, 
180-182); the embolus lamella is a huge 
flat scale with parallel sides of about the 
same visible area or larger than the small, 
narrow conductor ( Fig. 179 ) . The lamella, 
always easily seen, readily separates males 
from all other North American species. As 
no illustrations of the male palpus have 



been readily available, many males of this 
common species are misidentified in col- 
lections. 

Habits. Araneus trifolium is found in 
meadows and edges of fields making its 
web between herbaceous plants, often 
goldenrod {Soliclago sp. ), or shrubs, with 
the hub about 0.5-2 m off the ground. The 
web has about 20 spokes. The web and 
egg sac have been illustrated by Comstock 
(1912, 1940). During the recent years of 
drought the spiders disappeared completely 
from the fields around our house in Mas- 
sachusetts but reappeared with the first 
wet season in summer of 1967, to reach 
unusual abundance in 1968 and 1969. 
Kaston (1947) describes the egg sacs after 
Scheffer as "white, about an inch in diam- 
eter and delicate enough for the mass of 
several hundred yellow eggs to show 



Araneus diadematus group Orb-weavers • Levi 169 






oo 


u 


o 


3 


o 

J3 


' ■ 


1 


CO 


0) 




(- 


4^ 


o 


D 


T) 


n 




3 


CO 


<1) 


n 






(1) 

c- 


•~ 


o 


^ 


D 


-<t 


o 




t- 






O^ 














1 


o 


o 


-a. 


-=1. 






*- 


F 






ro 












(>v 


on 






~o 


-rr 




> 


^ 


oo 


o^ 










■ , 


*-- 






^ 


CO 
00 


oo 


oo 


o 

c 
(1) 


i/i 


: 


0> 


c> 


(U 


> 
















W1 


-> 


3 


3 


CN 
Ov 


D) 


:5 


O) 


O) 


D) 


u_ 


u. 


u. 


U- 


1 — 



o 

Q. 

>s 

O 

JH 

<u 

_c 

o 

E 

E 

o 
o 



o 

Si 

E 



E 
E 

IT) 

d 

o 



170 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



through. The young leave the egg sac in 
fall or spring." The egg sac is illustrated in 
Coinstock, 1912 ancri940. 

Distribution. Araneus trifolium is found 
from Alaska to southern California and 
northern Florida (Map 7). 

Araneus cavaticus (Keyserling) 
Plate 2; Figures 187-194; Map 8 

Epeira cavatica Keyserling, 1882, Verb. Zool. Bot. 
Ges. Wien, 31: 269, pi. 11, fig. 1, $. Female 
holotype from a Kentueky cave, Carter Co., 
Kentucky (Sanborn, 1874) in the Museum of 
Comparative Zoology, examined. 

Epeira cinerea Emerton, 1884, Trans. Connecticut 
Acad. Sci., 6: 302, pi. 33, fig. 10, pi. 35, figs. 
7, 8, $ . Female syntypes from Sable Chasm, 
New York, in the Museum of Comparative 
Zoology, examined. 

Aranea cavatica, — Comstock, 1912, The Spider 
Book, p. 470, figs. 487^89, 9 . Roewer, 1942, 
Katalog der Araneae, 1: 859. Kaston, 1948, 
Bull. Connecticut Ceol. Natur. Hist. Surv., 70: 
251, figs. 798-799, 9, i. 

Araneus cavaticus, — Bonnet, 1955, Bibliographia 
Araneorum, 2: 453. 

Note. Aranea sinistra F. P.-Cambridge, 
1904, Biologia Centralia-Americana, 2: 510, 
pi. 48, fig. 21, 9, from Omilteme, Mexico, 
may also be this species. The genitalia 
are similar but not identical. No specimens 
from intermediate areas were available. 
This species has been renamed A. sinistrella 
by Roewer, 1942, Katalog der Araneae. 

Description. Female. Carapace yellow- 
brown, darker anterior of thoracic de- 
pression. Sternum dark brown. Coxae light 
yellow. Legs yellowish brown with darker 
brown bands. Abdomen gray to brown 
dorsally with an indistinct folium and an- 
terior white mark ( Fig. 190 ) . Venter black 
between epigynum and spinnerets, with a 
white bracket on each side. Abdomen 
with two humps and covered by short hairs 
on dorsum, longer hairs on sides, and fine 
hair between humps. Total length of a 
specimen from Kentucky, 18 mm. Cara- 
pace, 7.3 mm long, 5.9 mm wide. First 
femur, 10.0 mm; patella and tibia, 11.8 
mm; metatarsus, 8.4 mm; tarsus, 2.7 mm. 



Second patella and tibia, 11.1 mm; third, 
6.0 mm; fourth, 9.6 mm. 

Male. Coloration like that of female, 
usually light. The tibia is not modified 
and the coxae lack hooks or spurs. A speci- 
men from Kentucky measured 15 mm in 
total length. Carapace, 7.6 mm long, 6.7 
mm wide. First femur, 13.5 mm; patella 
and tibia, 17.2 mm; metatarsus, 13.8 mm; 
tarsus, 3.4 mm. Second patella and tibia, 
15.0 mm; third, 8.4 mm; fourth, 12.5 mm. 

Variation. Females vai^v from 13-22 mm 
in total length, with the carapace 6.1-7.5 
mm long, 5.4—6.1 mm wide. Males vary 
from 10-19 mm in total length, with the 
carapace 6.9-8.7 mm long, 5.2-7.8 mm 
wide. 

There is little color or moiphological 
variation in this light-colored species. 

Diapiosis. The epigynum of the female 
of this large humpbacked species can be 
confused with that of A. norchnanni. In 
posterior view the epigynum has a median 
sclerotized triangular piece (Fig. 189), that 
of A. nordmanni encloses a heart-shaped 
depression (Figs. 89, 92). The males can 
be recognized by their giant size and long 
legs. The large size separates males from 
those of the closely related Pacific coast 
A. r^emma which has a somewhat similar 
embolus and terminal apophysis. 

Habits. A. cavaticus makes its large webs 
witli about 20 spokes (Plate 2) on bams, 
on porches, and bridges. One old col- 
lection from 1874 indicated that the spider 
was collected from under a natural bridge 
in Kentucky. It also has been rt^ported 
from under overhanging cliffs, probably 
its original habitat. One male was collected 
while being fed on by a female A. diade- 
matus in Ontario. Archer ( 1940, J. Ala- 
bama Acad. Sci., 12: 28) reports the species 
from caves in Alabama. 

Distribution. Araneus cavaticus is found 
from New Brunswick, eastern Ontario and 
Nova Scotia to Texas. The southernmost 
record is of a pair collected at Houston, 
Texas (Map 8). 



Araneus DiADEMATus GROUP Orb-weavers • Levi 171 




4 ' • ■' 



Araneus gemmoides 
D 







Map 8. Distribution of Araneus cavaticus (Keyserling), Araneus gemmoides Chamberiin and Ivie, Araneus gemmo (McCook), 
Araneus pima sp. n., and Araneus illaudatus (Gertsch and Mulaik). 



Araneus gemmoides Chamberiin and Ivie 
Plate 2; Figures 195-202; Map 8 

Epeira gemma, — Keyserling, 1892, Die Spinnen 
Amerikas, 4: 115, pi. 6, fig. 85, $. Not A. 
gemma McCook. 

Araneus gemmoides Chamberiin and Ivie, 1935, 
Bull. Univ. Utah, Biol. Ser., 2(8): 22, pi. 10, 
fig. 80, $ . Female holotype from Salt Lake 
City belonging to the Utah collection, kept at 



the American Museum of Natural History, 
examined. Bonnet, 1955, Bibhographia Araneo- 
rum, 2: 507. 

Aranea gemmoides, — Roewer, 1942, Katalog der 
Araneae, 1: 860. 

Araneus canmonis Schenkel, 1950, Verb. Naturf. 
Ges. Basel, 61: 65, $. Female lectotype, here 
designated, from Canmore, Banff National Park, 
Canada, in the Naturhistorisches Museum Basel, 
examined. NEW SYNONYMY. 



172 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



Description. Female from Wisconsin. 
Carapace light brown, sternum brown with 
a median light longitudinal streak. Coxae 
yellowish. Legs yellow-brown, with in- 
distinct darker bands. Dorsum of abdomen 
light grayish brown with almost no pat- 
tern at all. Venter with two longitudinal 
dark marks behind epigynum and two light 
marks in front of spinnerets. Total length, 
15 mm. Carapace, 4.6 mm long, 4.2 mm 
wide. First femur, 4.9 mm; patella and 
tibia, 5.9 mm; metatarsus, 3.7 mm; tarsus, 
1.5 mm. Second patella and tibia, 5.5 mm; 
third, 3.0 mm; fourth, 4.9 mm. 

Male from Idaho. Coloration almost like 
that of female. Abdomen with remains of 
outline of a folium and an anterior median 
dorsal light streak and humps large. Coxae, 
legs not modified. Total length, 8 mm. 
Carapace, 4.5 mm long, 3.5 mm wide. First 
femur, 5.7 mm; patella and tibia, 6.7 mm; 
metatarsus, 4.5 mm; tarsus, 1.5 mm. Second 
patella and tibia, 5.7 mm; third, 3.5 mm; 
fourth, 4.9 mm. 

Variation. Total length of females varied 
from 13-25 mm; carapace 6.3-8.5 mm long, 
5.0-7.5 mm wide. Total length of males 
5.4-7.9 mm; carapace 3.2-4.3 mm long, 
2.6-3.5 mm wide. 

This species, unlike A. gemma, shows 
little variation except in color. Sometimes 
there is a distinct folium, sometimes not, 
and there may be a white cardiac mark. 
Only on the Pacific coast where it is in 
contact with the very similar A. gemma are 
some specimens dark. A single British 
Columbia specimen was black like A. 
saevtis. 

Diagnosis. The very tiny epigynum with 
a triangular scape (Fig. 195) readily sepa- 
rates A. gemmoides from other species, 
except for some A. gemma which may be 
hybrids. Males have a small bulb barely 
larger than the tibia (Figs. 199, 200); these 
proportions separate males from A. gemma. 
The male of A. gemmoides is much smaller 
in size than that of A. cavaticus. 

Habits. The habitat of this species is 
similar to that of the eastern A. cavaticus. 



I ha\'e collected it from under eaves of 
houses and on barns in Wisconsin that 
accommodated its huge web with about 
20 spokes (Plate 2). But in the west it 
seems to have been collected in lodgepole 
(Pinus contorta) woods, and on the Pacific 
coast on houses. It has been found in 
natural situations in cave entrances, rocky 
ledges, and one record is from Manzanita 
chaparral, 1100 m, San Gabriel C, Los 
Angeles Co., California (L. Pinter). 

The egg sac of A. gemmoides made in 
the laboratoiy is a fluffy, loosely woven 
sphere of orangish silk, 2 cm in diameter. 
A color illustration can be found in 
Gertsch ( 1949, American Spiders, Van 
Nostrand). 

Distribution. Araneus gemmoides is 
found from British Columbia to Wiscon- 
sin, south to Missouri and Arizona ( Map 
8). It appears rare along the Pacific coast. 
Isolated records are: Michigan, Rose Lake, 
Clinton Co. (D. E. Bixler ); Silver Hill, 
Alabama, Sept. 1945 (S. Nelson); and San 
Lorenzo Island [Baja California]. 

Araneus gemma (McCook) 
Figures 203-215; Map 8 

Epeira gemma McCook, 1888, Pioc. Acad. Natur. 
Sci. Philadelphia, p. 193, figs. 1, 2, 2 . Female 
lectotype here designated from California in 
the Philadelphia Academy Natural Sciences, 
examined. 1894, American Spiders, 3: 182, 
pi. 9, figs. 1, 2, pi. 10, fig. 6, 5, S. 

Araneus gemmtis, — Chaniberlin and Ivie, 1935, 
Bull. Univ. Utah, Biol. Sen, 2(8): 21, pi. 10, 
fig. 79, 5. 

Araneus pinis Chaniberlin and Ivie, 1935, Bull. 
Univ. Utah, Biol. Ser., 2(8): 22, pi. 10, fig. 
81, $ . Female holotype from Fillmore, Cali- 
fornia, in the University of Utah collection 
kept at tlie American Museum of Natural 
History, examined. NEW SYNONYMY. 

Araneus gemma, — Bonnet, 1955, Bibliographia 
Araneorum, 2: 506. 

Note. As pointed out first by Chamber- 
lin and Ivie, McCook's description is a 
composite. The two syntypes are different 
species. One is clearly the common species 
in the San Diego, California, area, the 
type locality. The otlier is probably the 



Araneus diadematus group Orb-weavers • Levi 173 




o 

E 



CN 
CN 



o 


0) 


CN 


i- 




o 




D 




a 






0) 


o 


E 


F 


o 


<u 


-a 


U- 


J3 




o 


o 






(U 


CN 


o 




E 


._ 



t^ 




o 








o^ 




CN 








' — 
















"5 








"5 












s 




o 








o 




—J 








— ' 




o6 








NO 
On 




o 

(N 












"o 


d 

D 






"o 






U 






^ 


Q. 


c 


.4_ 






c 


o 


OJ 


3 






(U 


u 


> 


o 






> 


, 




-C 








13 


r-s! 








id 


O 


o 

CN 


^ 




















^ 


U-1 

o 


Nit 






E 




CN 


CN 






D 


Cvi 


^ 




6 






O 
CN 


o 


d 


E 




CD 

'd. 
m 


Q. 
D 


CM 


o 


E 
dj 

CD 


E 
E 


rv! 


u 


c 


^ 


<t 


On 


^ 


^ 






Ln 








X 




1 


1 




H 


vi 


On 


"B 


CN 




c 
dJ 






U 




a! 


E 




.— 


' ' 




T3 


o 

-a 


ai 


o 

CN 


E 


3 


o 


> 




D 


"o 


E 


D 




»/i 


^ 


J3 


E 




-o 


D 


cn 


E 


01 


^ 


c 
o 


o 


a 


LU 


CD 


_□ 


m 








c 


E 




-^' 


^ 


E 


^ 


LU 


On' 
O 


CN 




E 


<u 

J3 


CN 


CN 

1 


CO 





U1 

o 


E 

D 


O 


CO 




u 


CN 


o 


CN 


d 


^ 


CN 




-D 


VI 




O 

CN 


,J^ 


"o 


CJ 

d> 
Q. 


Q- 







O 


> 


3 


o 


"5 


u 


3 


O 


5 
E 




(J 


Cvi 


"o 


E 


> 





CN 


E 

D 
C 

>N 


^ 


3 

<u 

c 


o 
o 

CN 


E 

E 

dl 




e' 


D 






s 


Q. 


E 


^5 


"5 


^' 




LU 


to 




lU 




, 




d 


CN 


s 


-^' 


CN 


fNv^ 


VI 


O 












CN 




CN 




CN 


o 


J. 


^ 


cJ, 


3 


1 
U1 


D 


o 




o 


Q- 




u 






CN 


O 


CN 


T3 


irt 




t/i 


£2. 


VI 


c 


o 




OJ 




d) 




z> 


D 


d) 


3 


dJ 

N 


O) 


O 

a. 


D) 


O 


CD 


u_ 


U- 


2 


Ll_ 


kl 



174 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



same as what is called A. pima here. Other 
specimens in the McCook collection were 
A. gemmoides. The specimen which may 
be A. pima was not chosen lectotype be- 
cause it either has the wrong collecting 
locality or comes from the very border of 
its range. Also the specimen appeared to 
be an atypical A. pima. 

To favor stability of names, the other 
specimen was chosen lectotype even though 
as a result A. pirus will have to be synony- 
mized. The range of the species whose 
name is now A. gemma is essentially that 
given by McCook and also McCook's de- 
scription matches this species. Araneus 
gemma of Chamberlin and Ivie is certainly 
this species although the specimens could 
not be located. The name A. gemma has 
at times been used for A. pivm on collect- 
ing vials only. The name gemma is a noun 
in apposition; its ending does not change 
with the gender of the genus (Bonnet, 
1955). 

Description. Female from California. 
Carapace maculated brown with some 
gray pattern. Legs yellowish, indistinctly 
banded brown. Abdomen brownish with a 
median longitudinal white line, or line 
absent. Venter with a black band enclosed 
by white brackets. Abdomen with large 
humps (Fig. 210). Total length, 12 mm. 
Carapace, 5.3 mm long, 4.6 mm wide. First 
femur, 5.9 mm; patella and tibia, 8.2 mm; 
metatarsus, 5.0 mm; tarsus, 1.9 mm. Second 
patella and tibia, 8.0 mm; third, 4.6 mm; 
fourth, 6.8 mm. 

Male. Carapace, sternum, legs yellow- 
brown. Abdomen with a folium and an- 
terior median white mark. Venter gray 
with a white bracket on each side. First 
coxa with a small tubercle on distal margin. 
Second tibia not modified. Total length, 
8 mm. Carapace, 4.2 mm long, 3.3 mm 
wide. First femur, 5.9 mm; patella and 
tibia, 7.0 mm; metatarsus, 4.6 mm; tarsus, 
1.7 mm. Second patella and tibia, 6.0 mm; 
third, 3.3 mm; fourth. 4.9 mm. 

Variation. Females vary 9-19 mm in 
total length; carapace 4.4-6.4 mm long. 



3.8-6.1 mm wide. Males vary 5.8-8.5 mm 
in total length; carapace 3.1-4.5 mm long, 
2.5-3.6 mm wide. 

The color varies in alcoholic specimens 
from light gray to almost black, some have 
two bars on the venter. Hardly two speci- 
mens have similar epigyna. It is believed 
that the species hybridizes with Araneus 
gemmoides and introgression is taking 
place. The male palpi, similar to those of 
A. gemmoides except for proportions, were 
not carefully studied for variation. (See 
introduction under species problems.) 

Diagnosis. The short scape of the epigy- 
num differs from that of related species in 
having a median ridge (Figs. 203, 205, 
207, 208). But some females cannot be 
separated from A. gemmoides. (See Vari- 
ation above and introduction under species 
problems.) The males are much smaller 
in size than those of the eastern A. cavati- 
cus. The palpus has a larger bulb and 
smaller tibia (Figs. 211, 212) than the 
palpus of A. gemmoides. The upper sur- 
face of the embolus differs (Figs. 211, 213, 
214) from that of A. gemmoides and the 
terminal apophysis is more pointed. 

Note. McCook when describing A. 
gemma must have noted the problems of 
variation and applied one name to A. 
gemmoides and A. gemma. Presumably 
he could not readily separate the specimens 
which he had from California. Chamberlin 
and Ivie having smaller collections at hand 
used gemma for some specimens (Cham- 
berlin and Ivie, 1935, fig. 79) and called 
others A. pirus (Chamberlin and Ivie, 
1935, fig. 81), a reasonable judgment if 
only very few specimens are examined. 

Habits. Little is known of the habits of 
A. gemma, although they are probably the 
same as those of A. cavaticus and A. gern- 
moides. One specimen was collected on a 
redwood (Sequoia sempervirens) trunk, 
San Jose, California. One from a porch at 
Walla Walla, Washington, others from 
Moscow, Idaho, came in the same col- 
lection with A. gemmoides. One A. gemma 
was collected from a hole in an olive tree 



Araneus diadematus group Orb-weavers • Levi 175 




176 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



(Olea europea) feeding on a ruby-crowned 
kinglet (Regulus calendula) which was 
caught in its web (Pasadena, Cahfoniia, 
5 November, 1942, Mrs. H. Michener). 

Distribution. Araneus gemma is found 
from southern Alaska to southern Cali- 
fornia; the northernmost collections are 
from Ketchikan, Alaska, the easternmost 
from Bigfork, Montana (Map 8). 

Araneus pima sp. n, 

Plate 1, 4; Figures 218-232; Map 8 

Holotype. Female holotype from Madera 
Canyon, Santa Rita Mountains, Pima Co., 
Arizona, 29 July 1958 (A. Ross) in the 
American Museum of Natural History. The 
specific name is a noun in apposition after 
the type locality. 

Description. Female from Arizona. Cara- 
pace brown. Sternum brown with median 
area lighter. Legs brown. Abdomen gray- 
brown, without folium, the venter with a 
pair of parallel longitudinal black bars 
surrounded by white (Plate 4, Fig. 227). 
Humps very large. Abdomen covered by 
sparse long setae. Total length, 20 mm. 
Carapace, 8.0 mm long, 6.4 mm wide. First 
femur, 7.8 mm; patella and tibia, 9.8 mm; 
metatarsus, 6.5 mm; tarsus, 2.2 mm. Second 
patella and tibia, 9.6 mm; third, 5.5 mm; 
fourth, 8.9 mm. 

Male from Arizona. Coloration like that 
of female except that legs show some band- 
ing and abdomen has a folium and an- 
terior median longitudinal white line. 
Coxae and second tibia arc not modified. 
Total length, 9 mm. Carapace, 5.5 mm 
long, 4.3 mm wide. First femur, 7.0 mm; 
patella and tibia, 10.0 mm; metatarsus, 5.9 
mm; tarsus, 1.9 mm. Second patella and 
tibia, 7.6 mm; third, 4.2 mm; fourth, 6.4 
mm. 

Variation. The largest female measured 
27 mm in total length; the carapace was 
not measured. Another male measured 5.2 
mm in total length, carapace, 2.5 mm long, 
2.0 mm wide. Although what are believed 
to be tips of A. illaudatus emboli are found 



at times in the groove of the epigynum, the 
species do not seem to hybridize. There 
is not much variation except in length of 
scape. 

Diagnosis. The flat rounded scape of 
the epigynum (Figs. 218, 221, 224) with a 
deep wide groove under it (Figs. 220, 223) 
distinguishes females of A. pima from A. 
gemmoides and A. gemma. The males are 
distinguished from A. gemma by the shape 
of the median apophysis, by having the 
median spine joined with the distal one, 
and by the truncate embolus (Figs. 228, 
232). The more blunt terminal apophysis 
(Figs. 228, 229) and the shape of the 
median apophysis with the spines joined 
distinguishes A. pima from A. illaudatus. 

Habits. Araneus pima has been collected 
under eaves of buildings of the South- 
western Research Station, near Portal, Ari- 
zona, under a bridge in New Mexico, along 
a trail in the Grand Canyon, and in cave 
enti-ance of O.T.L. cave north of Alpine, 
Texas. 

The eggs are in an oval mass 2 cm long, 
1.8 cm wide, 1 cm thick. About 700 eggs 
were estimated on the surface of the clump 
of eggs, which must have contained more 
than a thousand. A loose pink woolly web 
4-5 cm- surrounded the eggs, quite similar 
to the egg sac of A. gemmoides (Plate 1). 

Note. In some collections specimens of 
this species had been incorrectly deter- 
mined as A. gemma. It is not A. gemma of 
McCook nor of Chamberlin and Ivie ( 1935, 
fig. 79). 

Distribution. Araneus pima is found from 
eastern California to Utah and Arizona. 
Localities at the border of the range are: 
Gateway, Oregon, Kerrville, Texas (Map 
8). Female and male para types from the 
type locality have been collected. 

Araneus illaudatus (Gertsch and Mulaik) 
Figures 233-240; Map 8 

Armiea illaiidata Gertsch and Mulaik, 1936, Amer. 
Mus. Novitates, 863: 19, figs. 36, 37, $. Male 
holotype from Edinburg, Texas, in the American 
Museum of Natural History, examined. Archer, 



Araneus diadematus group Orb-weavers • Levi 177 



1951, Amer. Mus. Novitates, 1487: 36, fig. 
74, S (not fig. 67 2 ). 

Note. Archer ( 1951 ) matched the wrong 
female to the male described previously 
as A. illaudatus. Archer's female is an A. 
nordmanni and as Archer correctly points 
out is quite different from females of A. 
cavaticus group. 

Description. Female from Arizona. Cara- 
pace yellowish with dark brown mottling. 
Sternum black, coxae yellowish, legs 
banded yellowish and dark brown. Dorsum 
of abdomen with two rows of transverse 
black marks pointing posteriorly towards 
the middle on a mottled background (Fig. 
236). A median dorsal longitudinal line of 
white spots not always present. Venter 
with a black band between pedicel and 
spinnerets containing a pair of prominent 
white spots side by side and sometimes a 
smaller pair of white spots. Abdomen with 
two humps. Total length, 11 mm. Cara- 
pace, 3.8 mm long, 3.4 mm wide. First 
femur, 4.2 mm; patella and tibia, 5.5 mm; 
metatarsus, 3.7 mm; tarsus, 1.4 mm. Second 
patella and tibia, 5.0 mm; third, 3.0 mm; 
fourth, 4.9 mm. 

Male holotype. Carapace brown. Ster- 
num brown. Legs yellow-brown. Abdo- 
men yellow-brown. Dorsum with distinct 
folium, anterior border has a white cross. 
Venter with a pair of white spots side by 
side. Coxae and second tibia not modi- 
fied. The abdomen has two distinct humps. 
Total length, 3.6 mm. Carapace, 2.0 mm 
long, 1.4 mm wide. First femur, 2.5 mm; 
patella and tibia, 2.8 mm; metatarsus, 1.8 
mm; tarsus, 0.9 mm. Second patella and 
tibia, 2.3 mm; third, 1.3 mm; fourth, 2.0 
mm. 

Diagnosis. The female scape is con- 
stricted at its base and almost diamond 
shaped (Fig. 233), quite different from 
that of the related A. pima (Figs. 218, 221, 
224). The male palpus has the spines of 
the median apophysis farther apart and the 
terminal apophysis more pointed and 
twisted (Figs. 238, 239) than that of 



Araneus pima. The cap of the embolus is 
the longest of any Araneus species, longer 
than the embolus (Fig. 240). 

Habits. One female collected in the 
Chiricahua Mts., Arizona, at 7500-9000 feet 
[2300-2900 m] had its web on brush under 
pines in August. 

Distribution. Araneus illaudatus is found 
from western Texas to Arizona (Map 8). 

REFERENCES 

Archer, A. F. 1951a. Studies in the orbweaving 
spiders (Argiopidae) 1. Amer. Mus. Novi- 
tates, 1487: 1-52. 

. 1951b. Studies in the orbweaving spiders 

(Argiopidae) 2. Amer. Mus. Novitates, 1502: 
1-34. 

Bonnet, P. 1955-1961. Bibliographia Araneorum. 
Toulouse, Vols. 2, 3. 

Brown, W. L., and E. O. Wilson. 1956. Char- 
acter displacement. System. ZooL, 5: 49-64. 

Chamberlin, R. v., and W. Ivie. 1935. Miscel- 
laneous new American Spiders. Bull. Univ. 
Utah, Biol. Ser., 2(4): 1-79. 

. 1944. Spiders of the Georgia region of 

North America. Bull. Univ. Utah, Biol. Ser., 
8(5): 1-267. 

CoMSTOCK, J. H. 1910. The palpi of male spiders. 
Ann. EntonioL Snc. Amer., 3: 161-185. 

Gertsch, W. J. 1964. The spider genus Zygiella 
in North America (Araneae, Argiopidae). 
Amer. Mus. Novitates, 2188: 1-21. 

Grasshoff, M. 1964. Die Kreuzspinne Araneus 
pallidas — ihr Netzbau und ihre Paanmgs- 
biologie. Natur. Mus., 94: 305-314. 

. 1968. Morphologische Kriterien als 

Ausdruck von Artgrenzen bei Radnetzspinnen 
der Subfamilie Araneinae (Arachnida: Ara- 
neae: Araneidae). Abhandl. Senckenbergi- 
schen Naturforsch. Ges., 516: 1-100. 

Helsdingen, p. J. VAN. 1965. Sexual behavior 
of Lepthyphantes lepiosus witli notes on the 
function of the genital organs. Zool. Med., 
41: 15-42. 

International Code of Zoological Nomen- 
clature. 1964. Int. Trust Zool. Nomencl., 
London. 

Levi, H. W. 1957a. The spider genera Enoplo- 
gnatha, Theridion and Paidisca in America 
north of Mexico (Araneae, Theridiidae ) . Bulb 
Amer. Mus. Nat. Hist., 112: 1-123. 

. 1957b. The spider genera Cmstulina and 

Steatoda in North America, Central America 
and the West Indies (Araneae, Theridiidae). 
Bull. Mus. Comp. Zool., 117: 367-424. 

. 1968. The spider genera Gea and Argiope 



178 Bulletin Museum of Comparative Zoology, Vol. 141, No. 4 



in America (Araneae: Araneidae). Bull. Mus. 

Comp. Zool., 136: 319-352. 
.(in press). Problems in the reprodnctive 

physiology of the spider palpus. Bull. Mus. 

Natl. Hist. Natur. Paris. 
— , AND L. R. Levi. 1961. Some comments 

on Walckenaer's Names of American Spiders, 



based on Abbot's drawings. Psyche, 68: 

53-57. 
WiEHLE, H. 1931. Araneidae. In F. Dahl, Die 

Tierwelt Deutschlands, 23(6): 1-136. 
. 1963. Beitriige zur Kenntnis der deut- 

schen Spinnenfauna III. Zool. Jahrb. Abt. 

System., 90: 227-298. 



Araneus diadematus group Orb-weavers • Levi 179 



INDEX 



Valid names are printed in italics. Page numbers 
refer to main references, starred page numbers to 
illustrations. 

ahine, Araneus, 161*, 162 
andrewsi, Aranea, 146 
andrewsi, Araneus, 145*, 146 
angulata, Aranea, 142 
angulatus, Araneus, 141*, 142 
approximata, Epeira, 167 
Aranea, 133 
Araneus, 133 
aureola, Epeira, 167 



bicentenaria, Aranea, 143 
bicentenaria, Epeira, 143 
hicentenarius, Araneus, 139*, 
Burgessia, 133 



143, 145^ 



candidans, Epeira, 167 
canmorus, Araneus, 171 
cavatica, Aranea, 170 
cavatica, Epeira, 170 
cavaticus, Araneus, 169*, 170 
cinerea, Epeira, 170 
corticaria, Aranea, 158 
corticaria, Epeira, 158 
corticarius, Araneus, 158, 159* 

darlingtoni, Aranea, 152 
denningi, Aranea, 158 
diadema, Epeira, 147 
diademata, Aranea, 147 
diadematus, Araneus, 147, 149*, 15 

Epeira, 133 
Euaranea, 133 

gemma, Araneus, 172, 173* 
gemma, Araneus, 176 
gemma, Epeira, 171, 172 
gemmoides, Araneus, 171, 173* 
gemmus, Araneus, 172 
gigas, Aranea, 156 
gigas, Araneus, 143 
gigas, Epeira, 143 
gosogana, Aranea, 167 
groenlandicolus, Araneus, 164, 165"* 

illaudata, Aranea, 176 
illattdatus, Araneus, 175*, 176 



169" 



incestifica, Epeira, 158 
insularis, Epeira, 156 
iviei, Aranea, 162 
iviei, Araneus, 161*, 162 

jaspidata, Epeira, 167 

kisatchia, Aranea, 143 

manitobae, Aranea, 164 

mannorea, Epeira, 156 

marmoreus, Araneus, 141*, 155*, 156, 159*, 169* 

Neopora, 133 

nigra, Epeira, 148 

nordmanni, Aranea, 150 

nordmanni, Araneus, 150, 151*, 153*, 155* 

nordmanni, Epeira, 150 

obesa, Epeira, 156 

pima, Araneus, 140*, 175*, 176 
pirus, Araneus, 172 
pseudomelaena, Aranea, 152 

quadratus, Araneus, 164, 165* 

raji, Aranea, 156 
raji, Epeira, 156 
reaumurii, Aranea, 164 

sachimau, Aranea, 162 

saeva, Aranea, 148 

saeva, Epeira, 148 

saevus, Araneus, 141*, 148, 149*, 151* 

santarita, Aranea, 150 

santarita, Araneus, 149*, 150 

silvatica, Epeira, 148, 150 

sinistra, Aranea, 170 

sinistrella, Aranea, 170 

solitaria, Aranea, 148 

solitaria, Epeira, 148 

trifoliimi, Aranea, 167 
trifolium, Araneus, 167, 169* 
trifolium, Epeira, 167 
tusigia, Aranea, 156 

vulpecula, Epeira, 167 

loashingtoni, Araneus, 155*, 159*, 160 

yukon, Araneus, 165*, 166 






in OF THE 



SuLU 
Museum of 
Comparative 
Zoology 



Evolutionary Relationships of Some 
South American Ground Tyrants 



W. JOHN SMITH AND FRANCOIS VUILLEUMIER 



HARVARD UNIVERSITY VOLUME 141, NUMBER 5 

CAMBRIDGE, MASSACHUSETTS, U.S.A. 4 MARCH 1971 



PUBLICATIONS ISSUED 

OR DISTRIBUTED BY THE 

MUSEUM OF COMPARATIVE ZOOLOGY 

HARVARD UNIVERSITY 

Bulletin 1863- 

Breviora 1952- 

Memoirs 1864-1938 

JoHNSONiA, Department of Mollusks, 1941- 

OccAsiONAL Papers on Mollusks, 1945- 



Other Publications. 

Bigelow, H. B., and W. C. Schroeder, 1953. Fishes of the Gulf of Maine. 
Reprint, $6.50 cloth. 

Brues, C. T., A. L. Melander, and F. M. Carpenter, 1954. Classification of 
Insects. $9.00 cloth. 

Creighton, W. S., 1950. The Ants of North America. Reprint, $10.00 cloth. 

Lyman, C. P., and A. R. Dawe (eds.), 1960. Symposimn on Natural Mam- 
malian Hibernation. $3.00 paper, $4.50 cloth. 

Peters' Check-list of Birds of the World, vols. 2-7, 9, 10, 12-15. (Price list on 
request. ) 

Turner, R. D., 1966. A Survey and Illustrated Catalogue of the Teredinidae 
(Mollusca: Bivalvia). $8.00 cloth. 

Whittington, H. B., and W. D. I. Rolfe (eds.), 1963. Phylogeny and Evolution 
of Crustacea. $6.75 clotli. 

Proceedings of the New England Zoological Club 1899-1948. ( Complete sets 
only. ) 

Publications of the Boston Society of Natural History. 



Authors preparing manuscripts for the Bulletin of the Museum of Comparative 
Zoology or Breviora should send for the current Information and Instruction 
Sheet, available from Mrs. Penelope Lasnik, Editor, Publications Office, Museum 
of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138, 
U.S.A. 



© The Presidenf and Fellows of Horvord College 1971. 



EVOLUTIONARY RELATIONSHIPS OF SOME 
SOUTH AMERICAN GROUND TYRANTS^ 

W. JOHN SM1TH2 AND FRANCOIS VUILLEUMIER^ 



PREFACE 

The Tyrannidae comprise about one 
out of ten of the approximately 2900 
species of birds hsted from the South 
American continent by de Schauensee 
(1966). Parallehng this large representa- 
tion is ecological versatility; tyrant fly- 
catchers occur in every major avian habitat 
from lowland rain forest to high Andean 
paramos. As a corollary to this radiation, 
many Tyrannidae have ecological char- 
acteristics which are typical of totally un- 
related Old World avian groups. There 
are, among others, thrushlike tyrants, chat- 
like, shrikelike, warblerlike, and titmouse- 
like tyrants, in addition to flycatchers. 

The existence of the tyrannid radiation 
has been known to ornithologists for a 
long time. Yet a detailed evaluation of 
even a part of it and of the evolutionary 
processes involved has not been attempted. 
Failure to do so is due in a large measure 
to the provincialism (taxonomic as well as 
geographic) of many published accounts 
dealing with Tyrannidae. It is now evident 
that significant progress in our understand- 
ing of the radiation within the Tyrannidae 
depends on broad comparative surveys of 
geographic patterns and nonmoi-phological 
characters, especially ethological ones. 

Until recently, few detailed studies of 
behavior in tyrant flycatchers have been 



1 Published by a grant from the Wetinore Colles 
Fund. 

2 Department of Biology, University of Pennsyl- 
vania, Philadelphia, Pa. 19104. 

^ Biology Department, University of Massachu- 
setts, 100 Arlington Street, Boston, Mass. 02116. 



made. Among the significant exceptions, 
one might mention analvses of genera such 
as Mijiarchus (Lanyon,^ 1960, 1961, 1963, 
1965, 1967), Empidonax (Stein, 1958; John- 
son, 1963; Mumford, 1964), Tijranmis 
(Smith, 1966), and Smjomis (Smith, 1969 
and in press, 1 and 2). In spite of these 
analyses, we still know virtually nothing 
about a large proportion of the genera and 
species. Museum skins and occasional field 
observations by collectors are literally all 
the information available today on whole 
genera or groups of genera of Tyrannidae. 
During the past several years both of 
us have studied various tyrannids. We 
have approached our subject independently 
and with different outlooks. Smith has long 
been interested in the Tyrannidae from the 
viewpoint of behavior and evolutionary 
studies. Vuilleumier, on the other hand, has 
studied only some tyrannid genera in detail 
as part of an analysis of speciation phenom- 
ena in Andean birds. Despite our different 
approaches, we have found that we agree 
on all the more important points of re- 
lationships and evolution within a group 
of genera, most of which are called ground 
tyrants and live on or near the ground in 
open habitats of South America, especially 
in the Andes and Patagonia. These include 
Muscisaxicola, Agriornis, and Xolmis and 
appear closely related to Ochthoeca, Saij- 
ornis, and other genera which are, to vary- 
ing degrees, less terrestrial. 

bur results suggest a certain number of 
changes in the currently accepted taxo- 
nomic arrangement of these birds. More 
significantly, the results also suggest some 
ways in which flycatchers have radiated 



into some of the more open t\^pes of habi- 
tats of South America. 

Because we employed different methods, 
we shall present our information separately. 
Nevertheless, the results converge, and the 
two papers constitute what we consider 
to be a unit: together they represent an 
attempt toward a better understanding of 
the relationships and evolution of several 
genera of ground tyrants and allied genera. 



We do not pretend to have reached the 
final word about the systematics of these 
genera, and are aware of the gaps in our 
knowledge. Some of our conclusions are 
provisional. We hope, however, that the 
uncertainties will be an incentive for fur- 
ther research by other workers. 

W. John Smith 

Frangois Vuilleumier 



Chapter I. Generic Relationships and Speciation Patterns in Ochthoeca, 
Myiotheretes, Xolmis, Neoxolmis, Agriornis, and Alusc/sax/co/a 

FRANCOIS VUILLEUMIER 

Abstract. The open habitats of the high Andes, Patagonia, south-central Brazil and adjacent coun- 
tries are inhabited by several more or less closely interrelated genera of Tyrannidae. The adaptive radia- 
tion of these flycatchers into nonforest vegetation resulted in a variety of forms that are the ecological 
counterparts of species from other families (Turdidae, Laniidae) elsewhere. At one extreme of this radia- 
tion are birds of a rather generalized "flycatcher type" (e.g., Ochthoeca), while at the other are birds that 
have diverged considerably, and are now entirely terrestrial in habits and wheatearhke in morphology 
(e.g., Muscisaxicola) . 

In Part I, morphological and ecological attributes of the genera Agriornis, Neoxolmis, Xolmis, Pijrope, 
Muscisaxicola, Mtiscigralla, Myiotheretes, and Ochthoeca (as listed by de Schauensee, 1966) are reviewed. 
As a result of this analy.sis, it is suggested that (a) the genus Xolmis be enlarged to include also species 
previously placed in the genera Myiotheretes and Pyrope, (b) Neoxolmis be retained as a monotypic 
genus, (c) Xolmis murina be transferred to the genus Agriornis, and (d) tlie genus Muscigralla be made 
a subgenus of Muscisaxicola. Diagnoses of the genera Ochthoeca, Xolmis, Neoxolmis, Agriornis, and 
Muscisaxicola are given, together with descriptions of the recognized species-groups. 

In Part II, geographical variation, species limits, and interspecific relationships are analyzed in Xolmis, 
Neoxolmis, Agriornis, and Muscisaxicola. Active speciation can be detected in the genera Xolmis, Agri- 
ornis, and Muscisaxicola. One can find: (a) species with discontinuous geographical variation across 
ecological barriers (e.g., Xolmis fumigata, Muscisaxicola maculirostris), (b) geographically isolated taxa 
at the borderline between species and subspecies ( superspecies ) (e.g., Xolmis pernix, X. fumigata, and X. 
fuscorufa; Muscisaxicola juninensis and M. albilora), (c) instances of secondary contact involving mar- 
ginal overlap (in the Muscisaxicola alpina superspecies), or possible hybridization (in Xolmis striaticollis 
and Agriornis montana ) . 

The available data suggest tliat in this group of genera, speciation is now taking place chiefly in the 
high Andes, where geographical isolation occurs across several well-marked ecological barriers. Zones of 
secondary contact in other regions, now devoid of such barriers ( in northern Bolivia ) , indicate that geo- 
graphical isolation was once possible there. " 

Nevertheless, there is so much extensive sympatry between congeneric species, in the Andes ( e.g., in 
Muscisaxicola) as well as in the lowlands (e.g., in Xolmis) that the early stages of their radiation cannot 
now be detected. 

TABLE OF CONTENTS Part II: Variation and Speciation in the 

Genera Xolmis, Neoxolmis, Agriornis, and 

Introduction 182 Muscisaxicola 203 

Parti: The Genera and their Relationships —. 183. The Genus Xolmis 204 

Classification 185 The Genus Neoxolmis __ 208 

Analysis of Some Characters 188 The Genus Agriornis 208 

The Genus Ochthoeca 192 The Genus Muscisaxicola 215 

The Genus Xolmis 193 Discussion 224 

The Genus Neoxolmis 197 Presently Actixe Speciation 226 

The Genus Agriornis 198 Older Speciation Patterns 229 

The Genus Muscisaxicola 200 Literature Cited 230 

Bull. Mus. Comp. ZooL, 141 (5) : 181-232, March, 1971 181 



182 Bulletin Mu.seinii of Comparative Zoology, Vol. 141, No. 5 



INTRODUCTION 

As was explained in the preface, this 
paper, toG;ether with the one by Smith, 
must be considered as an attempt to de- 
scribe the adaptive radiation of some 
groups of tyrant flycatchers into open, non- 
forested habitats of South America, espe- 
cially the grasslands and scrub of the high 
Andes. 

My study of several genera of Tyranni- 
dae was part of an analysis of speciation 
phenomena among Andean and Patagonian 
birds. Geographical variation and inter- 
species relationships were investigated in 
each genus selected for this work. I ana- 
lyzed especially, first, patterns of range 
discontinuities within the distribution of 
species, and second, areas of secondary 
contact between taxa at, or slightly below, 
the le\'el of species. Knowledge of these 
phenomena gives us insight about the 
course of speciation in these birds ( Vuilleu- 
mier, 1969b). This original scope has been 
enlarged here to include a discussion of the 
probable relationships of these birds at the 
genus level. 

This paper, then, is divided into two 
parts. In the first, the taxonomie and 
evolutionaiy relationships of several genera 
of bush and ground tyrants are examined, 
and a classification of these birds slightly 
at variance with that in current use (e.g., 
de Schauensee, 1966) is suggested. In the 
second part, evolutionaiy phenomena at 
the species level are discussed in the genera 
Xohms (which here has been enlarged to 
include MyiotJwrete.s and Pyrope), Neoxol- 
mis, A'yrionii'i (includes murina, a species 
formerly put in Xohnis), and Muscisaxicola 
( includes Muscigralla ) . 

This paper is the third in a series of 
speciation studies on Andean birds (see 
Vuilleumier, 1968; 1969a). 

ACKNOWLEDGMENTS 

I am deeply indebted to Ernst Mayr for 
his advice and stimulation during my 
studies of speciation, and for having read 



and criticized an earlier version of this 
paper. The manuscript benefited greatly 
from the numerous suggestions of Beryl S. 
Vuilleumier. I wish to thank sincerely Roy 
F. Steinbaeh (Cochabamba), Germain 
Broquet (Quito), ITannes Vogt (La Paz). 
Pablo Legname (Tucuman), Maria and 
Hans-Wilhelm Koepcke (Lima), and Mario 
Ricardi ( Concepcion ) for their help during 
my field work in South America. Jan 
Pinowski (Warsaw), Alexander Wetmore, 
and Richard L. Zusi (Washington) kindly 
provided me with data on specimens, and 
O. S. Pettingill, Jr., generously lent me his 
field data on some Falkland Islands birds. 

I am grateful to the following persons 
who have helped me during my exami- 
nation of specimens in the collections under 
their care, or who have sent me specimens 
on loan: Villy Aellen (Museum d'Histoire 
Naturelle, Geneva), Dean Amadou and 
Charles O'Brien (American Museum of 
Natural History), James Bond (Academy 
of Natural Sciences of Philadelphia), Jean 
Dorst (Museum National d'Historie Natur- 
elle), Ian C. J. Galbraith (British Museum, 
Natural History), Antonio Olivares (Insti- 
tuto de Ciencias Naturales, Bogota), Claes 
C. Olrog (Instituto Miguel Lillo, Tucu- 
man), Raymond A. Paynter, Jr. (Museum 
of Comparative Zoology), W. H. Phelps, 
Jr., and the late W. H. Phelps (Caracas), 
the late R. A. Philippi (Museo Nacional, 
Santiago), Robert W. Storer (Museum of 
Zoology, University of Michigan), and 
Enist Sutter ( Naturhistorisches Museum, 
Basel). 

I acknowledge the generous financial 
assistance of the National Science Foun- 
dation (Grants G-19729 and GB-3167 to 
the Committee on Evolutionary Biology of 
Harvard University), the Bourse federale 
de voyages de la Societe helvetique des 
sciences naturelles (Switzerland), the So- 
ciety of the Sigma Xi, and the Frank M. 
Chapman Memorial Fund of the American 
Museum of Natural Histoiy. These grants 
made possible the study of bush and 



Evolution of Ground Tyrants • Smith and Viiilleumicr 



1S3 



ground tyrants during trips to South 
America in 1964, 1965, and 1967-1968. 

The illustrations of this paper were pre- 
pared by Margaret Estey, to whom I ex- 
press my thanks. Finally, I wish to thank 
Penelope Lasnik for her editorial assistance. 

PART I: THE GENERA AND THEIR 
RELATIONSHIPS 

The taxa analyzed below include eight 
genera recognized by de Schauensee 
(1966), who followed the unpublished 
work of Zimmer. They are Agriornis (4 
species), Neoxolmis (monotypic), Xohnis 
( 8 ) , Pyrope ( 1 ) , Muscisaxicola (11), 
Muscigralla (1), Myiotlieretes (6), and 
Ochthoeca (9). These eight genera are 
all more or less closely related to one 
another and probably represent a group 
of Tyrannidae that could conceivably be 
given suprageneric rank (e.g., tribe). I 
refrain from suggesting such a taxonomic 
procedure at the present time, however, 
because I have not made a sufficiently 
detailed analysis of other genera ( Tum- 
hezia, CoJoramphus, Ochthornis, Sayoniis, 
Pyrocephahts) that I believe to be more 
closely related to the above eight genera 
than their location in check-lists (Hell- 
mayr, 1927; de Schauensee, 1966) would 
imply. In this part, I will discuss the re- 
lationships of groups of species currently 
classified in the above eight genera, and 
will show that an alternative classifiqation 
to the one advocated by de Schauensee 
(1966) might better express the phenetic 
groupings of these species. 

The birds belonging to the above eight 
genera are distributed in the high Andes, 
from the timber line up to the upper limits 
of vegetation near the snow line, and from 
Colombia and Venezuela in the north to 
Patagonia in the south. In the lowlands 
and uplands of the continent, they occur in 
nonforest vegetation along the Pacific 
coastal plain and especially east of the 
Andes in Bolivia, Brazil, Paraguay, Argen- 
tina, and Uruguay. They are even found 



on the Falkland Islands off southern Pata- 
gonia. In view of this broad distribution in 
South America, it is interesting to point out 
that this group of genera does not occur in 
two other regions where the vegetation is 
predominantly nonforest, and where they 
might be expected to be found. One is the 
llanos of Venezuela and Colombia, and the 
other the high altitude grasslands of the 
mountains of southeastern Brazil. 

A first group of these genera {Xolmis, 
Pyrope, Myiotheretes, and Ochthoeca) can 
be called bush tyrants. A second group 
(Agriornis, Neoxolmis, Muscisaxicola, and 
Muscigralla) are ground tyrants: they 
usually live away from trees, in very open 
habitats such as scrub and grass asso- 
ciations. 

Ecological preferences vaiy at two levels 
in these eight genera. (A) Interspecific 
differences can clearly be seen in Xolmis, 
where X. ruhetra is largely terrestrial, in 
contrast to the other species, which are 
more arboreal, and in Ochthoeca, where 
species such as O. ciniiamomeiventris in- 
habit forested slopes, whereas the others 
live in open scrub (e.g., O. leucophrys). 
(B) Intraspecific variability is evident in 
Myiotheretes striaticoUis, which lives in 
wet woodlands in some parts of its range 
(e.g., in Ecuador), but in diy, xeric que- 
bradas in others (e.g., in Argentina). 

This ecological variability is, I believe, 
of evolutionaiy, and hence taxonomic, sig- 
nificance, and is inteqoreted here in terms 
of adaptive trends. In other words, these 
birds may have become adapted to in- 
creasingly open habitats during their evo- 
lutionary history. Such trends may have 
culminated in the adaptive radiation of 
Muscisaxicola, which has about ten species 
that live, often sympatrically, along the 
Andes in the paramo and puna grassland 
and scrub. Geologically speaking, these 
habitats are probably the most recent of 
any South American type of vegetation. 
The species of Muscisaxicola differ from 
most arboreal flycatchers in a series of 
characters that are probably all part of the 



184 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



adaptations (and concomitant evolutionary 
changes) to such extreme habitats: long 
legs, inconspicuous colors and patterns, and 
simplification of vocal manifestations. 

Unfortunately, ornithologists have so far 
done relatively little work that might en- 
lighten us about the modalities of this 
adaptive radiation. Part of the reason for 
this failure lies in the taxonomic treatment 
of these birds. It has been customary to 
keep the generic limits of taxa such as 
Agriornis, Xolmis, Muscisoxicola, and Och- 
thoeca relatively constant, because these 
genera contain species that resemble each 
other closely. But the more distinct or more 
intermediate taxa (usually single species) 
have either been placed in a monotypic 
genus or shifted back and forth from one 
genus to another. Such shifts can be made 
one way or the other, depending on which 
character a given author feels has more 
taxonomic weight as a "generic" character. 
Thus the species rufiventris is in some 
characters intermediate between Xolmis 
and Myiotheretes, but is distinct from 
either in other traits, and so was given 
generic rank (Neoxolmis) by Hellmayr 
(1927). The species pyrope differs in 
several ways from species ordinarily placed 
in Xolmis, and was consequently isolated 
in a monotypic genus Pyrope by Zimmer 
(MS; see de Schauensee, 1966). The spe- 
cies nifipennis was removed from Cnemor- 
chus and placed in Xolmis by Zimmer 
(MS). Finally, the species brevicouda has 
been isolated in a monotypic genus for so 
long that no one has asked whether this 
taxon was possibly related to others, and 
which ones they might have been. 

Part of the taxonomic problem just out- 
lined can be resolved by taking into con- 
sideration ecological and behavioral fac- 
tors, together with the more traditional 
ones of moipliology. The classification pre- 
sented in this paper is one attempt to do so. 
I suggest that the major ecological shift 
that has occurred in the evolutionary 
history of the bush and ground tyrants 
has been a change both from arboreal to 



terrestrial habits and from a relatively 
closed habitat (woodland) to a much 
more open one (steppe). The present 
geographical range of bush flycatchers in 
the nonforested Brazilian uplands, and 
various other nonforest vegetation for- 
mations of central and southern South 
America coincides in part with regions of 
great geological age that have not under- 
gone major changes in their position or 
structure. Indirect evidence (e.g., from 
fossil mammals) suggests that the vege- 
tation of such regions has been nonforest 
for a long time (i.e., since at least the mid- 
Tertiary). One may therefore suppose that 
the ecological shift has been chiefly in the 
direction of woodland to steppe, because 
the South American steppes (high Andean 
and Patagonian) appear to be much more 
recent than the central South American 
woodland formations. Whatever the actual 
direction of the shift, however, I have 
placed taxonomic weight on characters that 
appear to be correlated with it. 

Since a classification that includes only 
the categories genus and species is, in my 
opinion, not able to express satisfactorily 
the multiple hierarchical relationships of 
taxa, I have, as in previous systematic 
papers ( Vuilleumier, 1965, 1968,^ 1969a). 
made full use of the species-group and 
superspecies categories, which do not 
burden the formal nomenclature by ad- 
ditional names. I have also used subgenera 
in one case. 

The classification of the bush and ground 
tyrants constructed on the basis of an 
analysis of xariation in characters is out- 
lined below prior to the character analysis, 
in order to maintain throughout the paper 
as much uniformity as possible in the 
nomenclature of these birds. 

I am fully aware that a linear sequence 
of taxa cannot express all the possible re- 
lationships of these taxa. The sequence 
adopted here is therefore a compromise. 
The following conventions have been 
adopted. Genera having entirely or mostly 
arboreal and woodland-inhabiting species 



Evolution of Ground Tyrants • Smith and Vuilleumier 185 



are placed before genera having terres- 
trial and steppe-inhabiting species. Thus 
Ochthoeca and Xohnis are placed before 
NeoxoJmis, Agriornis, and Muscisaxicola. 
Ochtlweca precedes Xohnis because it may 
be more closely related to the two Andean 
species-groups of Xohnis (numbers 1 and 
2 in the list below) than to other genera. 
Agriornis is listed between Xohnis and 
Neoxohnis on the one hand, and Muscisaxi- 
cola on the other, because in some ways, 
it is intermediate between them. Within 
genera, the species-groups are listed with 
the Andean group ( s ) first, and the lowland 
one(s) next, except in Muscisaxicola, where 
this procedure would have relegated the 
maculirosfris species-group toward the end 
of the list, far away from the subgenus 
Muscigralla, with which it is more related 
than with other species-groups. Within 
species-groups, the species are listed either 
randomly if sympatric, or geographically 
from north to south if allopatric. 

Species-groups and superspecies have 
been named after the oldest name of any 
taxon within each. Species or semispecies 
considered to be component members of 
a superspecies are included in braces. 

Classification 

Genus Ochthoeca Cabanis, 1847 

1. cinnamomeiventris species-group 

O. cinnamomeiventris (Lafresnaye, 1843) 

2. diadema species-group 

O. diadema (Hartlaub, 1843) 

frontalis superspecies 
\ O. frontalis ( Lafresnaye, 1847 ) 
[ O. pulchella Sclater and Salvin, 1876 

3. rufipectoralis species-group 

O. rufipectoralis ( Lafresnaye and d'Or- 
bigny, 1837) 

4. oenuntJioides species-group 
O. fumicolor Sclater, 1856 

O. oenanthoides ( Lafresnaye and d'Or- 

bigny, 1837) 
leucophrys superspecies 
O. leucopJirt/s (Lafresnaye and d'Orbigny, 
1837) 
O. piiirae Chapman, 1924 

Genus Xolmis Boie, 1826 

1. fumigata species-group 

X. striaticollis (Sclater, 1853) 



fumigata superspecies 
X. pernix (Bangs, 1899) 
X. fumigata ( Boissonneau, 1840) 
X. fuscorufa (Slater and Salvin, 1876) 
(X. signata ( Taczanowski, 1874) is main- 
tained in this genus and species-group 
provisionally, pending critical examination 
of the one or two extant specimens. ) 

2. erijthropijgia species-group 

X. enjthropygia ( Sclater, 1853 ) 
X. rufipennis (Taczanowski, 1874) 

3. pijrope species-group 

X. ptjrope (Kitthtz, 1830) 

4. cinerea species-group 

X. cinerea (Vieillot, 1816) 
X. velata ( Lichtenstein, 1823) 
X. coronata (Vieillot, 1823) 
X. dominicana (Vieillot, 1823) 
X. irtipero (Vieillot, 1823) 

5. ruhetra species-group 

X. ruhetra (Bumieister, 1860) 

Genus Neoxohnis Hellmayr, 1927 

N. rufiventris (Vieillot, 1823) 
Genus Agriornis Gould, 1839 

1. montana species-group 

A. montarm (Lafresnaye and d'Orbigny, 

1837) 
A. alhicauda (Philippi and Landbeck, 

1863) 

2. hvida species-group 

A. Uvida (Kittlitz, 1835) 
A. microptera Gould, 1839 

3. murina species-group 

A. murina (Lafresnaye and d'Orbigny, 
1837) 

Genus Muscisaxicola Lafresnaye and d'Orbigny, 
1837 
subgenus Muscigralla Lafresnaye and d'Orbigny, 
1837 
M. hrevicauda Lafresnaye and d'Orbigny, 
1837 
subgenus Muscisaxicola Lafresnaye and d'Or- 
bigny, 1837 

1. maculirostris species-group 

M. maculirostris Lafresnaye and d'Orbigny, 

1837 
M. fluviatilis Sclater and Salvin, 1866 

2. macloviana species-group 

M. macloviana (Garnot, 1829) 
M. capLstrata (Burmeister, 1860) 

3. rufivertex species-group 

M. rufivertex Lafresnaye and d'Orbigny, 

1837 
alhilora superspecies 
M. juninensis Taczanowski, 1884 
I M. alhilora Lafresnaye, 1855 



I 



186 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



Muscisaxicola 




Ochthoeca 



Fig. 1. Three of the four groups based on differences in color pattern. Top: Muscisox/co/o ruHvertex, an example of group 
1 (unpotterned species with a contrasting crown patch). Center: Agriornis montana, an example of group 2 (species with 
streaked throat and patterned tail). Bottom: Ochthoeca rufipectoralis, an example of group 3 (species wih superciliary stripe 
and wing bars). 



Evolution of Ground Tyrants • Smith and Vuilleumier 187 












_j_. 


-^ 






O 








C 


■"■ 


-o 






?-5 








^ 


C 


c 






«n .(_ 








o 


^ 


^ 




-^ 


2 ■= 








-Q 

1 




2 








<u 


_^ 




>" 


1 


1 


D 


D 


u, Q 


■•" 


u 


X 


u 


>s 


>^ 


o 


o 






O 


t^ 


D 


O 


2 


1- 


M- 


o o 


_Q 


D) 


^ 


CD 


D) 


^ 


.. <D 



I 



ESS 



iB 



i i 



(U 

> 




a 

c 

■ 

X 





I/) 

"o 



>< 




_Q 



"^^^^^^^^1^ 






o c 

O .C 

-" i 

(1) T3 

C 






D 



O 



c 
D 



D 

U> 

C 



c *- 

(U 



O 



O 



r -D -a 

Q. =) -^ 

O - ^ 
cn D- ^=^ 



u> 






o 

E 



^ J2 j: 



-□ 
<U 

o 



O 

D. 



-a 



E o 

2 -d 

>* 0) 

5 E 

> o 






a «- 

D. O 



188 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



4. alpina species-group 
alpina superspecies 

{M. alpina (Jardine, 1849) 
M. cinerea Philippi and Landbeck, 1864 

5. alhifrons species-group 
alhifrous superspecies 

f M . alhifrons (Tschudi, 1844) 

I M. flavinncha Lafresnaye, 1855 

M. frontalis (Burmeister, 1860) 

Analysis of Some Characters 
Color and Paffern 

The color of the bush and ground tyrants 
varies from black and white {Xolmis 
irupero) to almost entirely dark smoky 
brown {Xolmis fumigota) or black (Och- 
tlweca cinnamomewentris) and to different 
shades of gray and brown, and, in some 
species, even greenish and yellowish. Some 
species are unpattemed, while others have 
striking patterns. 

This variability permits one to make 
four groups. (1) Largely unpattemed spe- 
cies having uniformly colored body, wing, 
and tail plumage. The outer rectrices 
usually have a pale margin. Some species 
have a conspicuous, others, a faint, crown 
patch of a yellowish, brownish, or reddish 
color; some species lack a crown patch 
altogether. This group includes all the 
species of Mtiscisaxicola plus Xolmis 
pyrope (Fig. 1). A partial exception is 
Muscisaxicola hrevicauda, which has a 
somewhat patterned rump and tail, al- 
though otherwise it has dull, unpattemed 
body plumage and a (concealed) crown 
patch. 

(2) Species having essentially unpat- 
temed body and wings ( save for pale outer 
edges of secondaries and coverts), and a 
white throat with dark streaks. The tail is 
patterned with white in some species, 
but is essentially unpattemed in others. 
No species has a cro\vn spot. This group 
comprises the five species here included in 
Agriornis (Fig. 1). 

(3) Species having either pattemed or 
unifomily colored body plumage, and a 
conspicuous white, buffy, or yellow super- 



ciliary stripe. Many species also have con- 
spicuous wing bars. The tail is essentially 
unpattemed, although a few species have 
a whitish or pale buffy edge to the outer 
rectrices. This group includes the species 
of Ochthoeco (Fig. 1). 

(4) Species having pattemed wings 
(often the pattern is concealed), and either 
pattemed or unpattemed body plumage 
and tail. The underwing pattern is veiy 
uniform. The distal one-third to one-half of 
the primaries is black or dark brown, the 
basal part rufous, white, or buff (Fig. 2). 
The exceptions are a white wing tip and 
black basal part {Xolmis clominicana) , and 
a pale rufous distal part of the primaries 
{Xolmis nihetra) (Fig. 2). All other spe- 
cies of Xolmis (but not X. pyrope, men- 
tioned above under 1) and Neoxolmis can 
be included in the patterned group. 

Attenuation of the Outer Primaries 

The two outer wing feathers vary from 
unmodified to highly attenuated at the tip. 
In most, and probably all, of the species 
having attenuated outer remiges, this at- 
tribute appears restricted to adult males. 
The possible functions of this modification 
during displays, for example, is still largely 
unknown. Wetmore (1926a: 302) indi- 
cated that Xolmis ruhetra made some sort 
of rattlelike noise in flight, which is most 
likely related to the attenuation of its outer 
primaries. Smith describes wing displays 
in several bush and ground tyrants. 

Two groups of birds can be made on the 
basis of variation in the degree of emargi- 
nation of the remiges. In the first (no 
attenuation) are all species of Ochthoeca, 
all species of Muscisaxicola, and several 
species of Xolmis (Fig. 3). In the second 
group are species having moderate to 
marked attenuation. They include all spe- 
cies of Agriornis { adult males only ) , several 
species of Xolmis, and Neoxolmis. Within 
Xolmis, variation in emargination appears 
to be species-specific as depicted in Figure 
3. 



Evolution of Ground Tyrants • Smith and Vuilleumier 189 




(1> 
a. 




o 

c 

E 



D 



O 
T3 




o 

c 




E 
o 




O 

c 
o 

c 
o 
E 




0) 

O 

>^ 

Q. 




c 
> 



c 
_o 

O) 

< . 

(1) 

§ b! 
^- .0 








o ^ 






X D 






2 Q 






2 






^ o 


«/> 




o/mis 
pecifi 




— 


— — ^^^' 





Q j^^ ^ 




X <" 


.y ^ 


a> 


C (1) 


^— ^ M 1***"^ _ 


, - o 


U Q. 


O -- — . — —---„>"- 


3 


i: ^-""^ 


1- 


M 


*/) 




. 


X 


Q. <= 


X 




Fig. 3. Attenuation of 
differences in attenuatio 



190 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



SIDE VIEW 



TOP VIEW 



Ochthoeca 
cinnamomeiventris 



Muscisaxicola 
alpina superspecies 




Xolmis 
cinerea 



rnis livida 




Fig. 4. Bill structure of Ochthoeca, Muscisaxicola, Xolmis, and Agriornis. O. cinnamomeiventris: short, brood bill, with long 
rictal bristles, correlated with insectivorous diet and wet, forested habitat. M. alpina-. thin, narrow bill, with short rictal 
bristles, correlated with insectivorous diet and dry, nonforested habitat. X. cinerea-. relatively long, broad bill, with long 
rictal bristles, correlated with partially insectivorous diet and moderately dense habitat. A. livida-. long, narrow, and sharply- 
hooked bill, with short rictal bristles, correlated with varied diet and nonforest habitat. Drawn to scale. 



6/// Structure 

The bill of the bush and ground tyrants 
is a very variable organ. Although the 
precise diets and feeding methods of most 
species are still very poorly known, it is 
possible to correlate positively certain 
characters of bill size and shape with food. 
Thus, small and insectivorous species 
{Ochthoeca and Muscisaxicola species) 
have relatively small bills, whereas larger 
species, of the genera Xolmis and Agriornis, 



having more diversified diets and occasion- 
ally or regularly capturing small verte- 
brates, have relatively powerful bills, often 
with a sharp hook at the tip (Fig. 4). 

Bill width varies considerably in the 
bush and ground tyrants. Species living in 
wetter and structurally denser habitats 
seem to have proportionately broader bills 
than do species living in dry and open 
habitats (Fig. 4). This correlation is more 
or less independent of size. Among open- 



Evolution of Ground Tyrants • Smith and Vuilleumier 191 



coimtiy species, small-sized Muscisaxicola 
and large-sized Agriornis all have relatively 
narrow bills. 

Shape of Tail 

Most species have a square or slightly 
forked tail. In some species, however, the 
tail is markedly forked (Neoxohnis riifi- 
ventris) or markedly rounded (Xolmis 
coronato). These differences appear to be 
only species-specific, and of little use for 
supraspecific groupings. 

Proporn'ons 

There is a tremendous amount of vari- 
ation in the proportions of wing, tail, tarsus, 
and culmen lengths within the bush and 
ground tyrants. In most instances, it is dif- 
ficult to establish correlations between 
proportions and other characters, or be- 
tween proportions and habits. The most 
terrestrial species, such as all species of 
Muscisaxicola, have proportionately long, 
thin legs. Arboreal species usually have 
proportionately shorter legs than do ter- 
restrial ones. More detailed discussions of 
the possible taxonomic significance of 
variation in proportions will be found in 
the generic accounts below. 

Size 

The flycatchers under consideration vaiy 
from very small ( Muscisaxicola brevi- 
cauda) to very large {Agriornis livida). 
Most of this variation appears to be inter- 
specific. In Muscisaxicola, some of the 
interspecific variability in size occurs be- 
tween pairs or among triplets of sympatric 
species living either in the same or in 
adjacent biotopes. These differences might 
represent character divergence permitting 
resource partitioning, whereas other dif- 
ferences, such as species-specific head or 
crown patterns, may permit recognition of 
conspecific individuals. 

In Agriornis, the two high Andean spe- 
cies {montana and alhicauda) are sym- 
patric over much of their respective ranges. 



They are very similar in pattern but differ 
in absolute size. The lowland species limda 
and microptera, although entirely allo- 
patric, differ conspicuously in size rather 
than in pattern. 

These two examples, in Muscisaxicola 
and Agriornis, show that ecological cor- 
relates of size differences may or may not 
exist. It seems difficult to assign any taxo- 
nomic significance to size variation among 
the bush and ground tyrants. 

Eye Color 

The iris of most species of bush and 
ground tyrants is dark brown or blackish, 
but that of Xohuis pyrope and of Xolmis 
cinerea is red. This color is conspicuous in 
the field. Such variation docs not seem to 
be of great taxonomic significance, except 
to suggest that X. ptjrope is relatively 
closely related to birds of the X. cinerea 
species-group. 

Nest Site 

When considered together with other 
characters, such as color pattern, pro- 
portions, habitat, and perhaps behavioral 
traits, nest site may be of help in assessing 
the interrelationships of some taxa. I have 
made use of such correlations especially in 
the case of Neoxolniis rufiventris and 
Muscisaxicola brevicauda (see below). In 
the genus Agriornis, however, the variation 
in nest site may be of less taxonomic 
significance. One of the three species living 
in high Andean steppes {montana) builds 
its nest on the ground in protected sites, 
whereas another {microptera) does not. 
Finally, A. livida, found in open woodlands 
and scrubby vegetation in lower zones 
along the Andes, builds its nest in trees. 
Data about nests and nest sites arc given 
by Smith. 

t-labifat Preferences and Distribution 

As I pointed out earlier, there is both 
intra- and interspecific variation in eco- 
logical preferences among the bush and 



192 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



ground tyrants. Further information is 
given by Smith. Three species occur in 
rather speciahzed habitats, and the pos- 
sible taxonomic significance of this fact 
needs to be examined. Xolmis pijrope ap- 
pears restricted to open woodlands and the 
edges of Patagonian forests dominated by 
trees of the genus Nothofagus. In other 
characters this species resembles Musci- 
so.xicola (absence of patterning) or XoJmis 
(proportions, eye color, arboreal habits, 
behavior). Since I have placed some weight 
on the transition from arboreal to terres- 
trial adaptations, I have maintained this 
species in Xolmis (in a distinct species- 
group) because it exhibits few of the more 
typical adaptations to terrestrial life found 
in other ground tyrants. 

Another ecological specialist is Neoxohnis 
rufiventris. This species occurs at the 
southern tip of South America, in open 
boglike grasslands or steppes with low 
Berheris shrubs. It apparently shares this 
habitat with no other species of the groups 
discussed in this paper. In its nest site, 
proportions, and general behavior, this 
species can only be considered a true 
ground tyrant. But in color pattern it 
resembles some of the arboreal species of 
Xolmis, especially of the Andean species- 
groups. Assuming that N. rufiventris has 
fully evolved to the terrestrial adaptive 
zone, I have maintained it in the monotypic 
genus erected by Hellmayr ( 1927 ) . 

The third species is Muscisaxicola hrevi- 
cauda. It lives in the lowlands of south- 
western Ecuador and northwestern Peru, 
in open habitats ranging from steppelike 
man-made biotopes (cotton plantations) to 
natural savannas and open woodlands. In 
habits it is partly terrestrial, like other 
Muscisaxicola, and partly arboreal, like 
other birds such as Xolmis. In pattern and 
behavior it is closest to Muscisaxicola, but 
its habitat is clearly denser. I have con- 
sidered it here the most aberrant member 
of the genus Muscisaxicola, and have 
placed it in a subgenus of its own. 



The Genus Ochthoeca 

Diagnosis 

A genus of about seven to nine species 
of small to medium-sized flycatchers living 
along the Andes in high altitude steppes, 
open woodlands, and, occasionally, wet 
montane forests. All forms of this genus 
have a relatively short but rather broad 
bill, and long and conspicuous rictal 
bristles (see Fig. 4). The outer primaries 
are not emarginated. Colors are usually 
browns and grays, but one species is largely 
greenish and another black and chestnut. 
All species have a head pattern with the 
loral spot and a superciliaiy stripe of a 
different color than the crown and cheeks 
(see Fig. 1). Most species have one or 
two wing bars (see Fig. 1), but this char- 
acter is geographically variable in at least 
one species (O. riifipectoralis). The under- 
wing color patterns, so striking in the genus 
Xolmis, are absent in Oclithoeca. Species 
of Ochthoeca have a characteristic field 
silhouette, with large head and convex 
forehead (see Fig. 1). 

Species Included 

I accept here the treatment of de 
Schauensee (1966: 338-340), who recog- 
nized nine species. These species can be 
arranged into four species-groups, de- 
scribed below. ( I believe that the following 
taxa, now placed rather far from Ochthoeca 
in the lists of Hellmayr [1927] and de 
Schauensee [1966], are closely related to, 
or even possibly congeneric with, Och- 
thoeca: Tu77ibezia salvini [Taczanowski, 
1877], CyOloramphus parvirostris [Danvin, 
1839], and Ochthornis littoralis [Pelzeln, 
1868].) 

1. The cinnamomeiventris species-group. 
The only species included, Ochthoeca cin- 
namomeiventris, is black with a white loral 
spot and supercilium. The underparts vary 
geographically from black to black and 
chestnut. Some populations of this species 
have conspicuously white axillary feather 
tufts. O. cinnamomeiventris has the broad- 



Evolution of Ground Tyrants • Smith and V uilleumier 193 



LENGTH ofCULMEN from SKULL 



mm 
















29n 








o 


Q> 






28- 
27- 








o 
99 


<5) 

q9 


o 


1 


26- 
25- 
24- 


DDD 


• 


• 
• 


A O 

o 
o 


o 




/ 


23- 


A 1 


«• 












22- 
21- 


A 


♦ 














1 

6 


1 
7 




1 
8 


1 

9 


1 
10 


1 
V 



o X .striaticollis 
^ X.pernix 

• X.fumigcta 

♦ X.fuscorufa 
° X.rufipennis 

^ X.erythropygio 



-, WIDTH ofCULMEN 
11mm at NOSTRILS 

Fig. 5. Length of culmen plotted against width of culmen in the Xolm/s lumigata species-group (1) and the X. erythropygio 
species-group (2). Note the absence of overlap in width of culmen between the two groups. 



est and flattest bill of any species of 
Ochthoeco (see Fig. 4). This species may 
be the most strictly forest-inhabiting species 
of Ochthoeca or, indeed, of the entire 
group of genera considered in this paper. 

2. The diadema species-group. This 
group contains Ochthoeca diademo, and a 
superspecies that includes O. frontalis and 
O. puIcheUa. All three species have a 
yellow loral spot, and a yellow or white 
supcrcilium. The upper parts vaiy from 
dark brown (in pulcliella) to greenisli (in 
diadema). The species of this group are 
forest birds, but they occupy relatively 
open biotopes, such as clearings, rather 
than the forest interior. 

3. Tlie rufipectoralis species- group. Och- 
thoeca rufipectoralis, the only species of 
this group, has a buffy supcrcilium and 
patterned underparts, with contrasting 
rufous chest and pale gray lower abdomen. 
A similar pattern is not found in any other 
species of the genus. O. rufipectoralis lives 
in more open habitats than birds of the 
first two species-groups. 



4. The oenanthoides species-group. The 
four species I include in this group (O. 
fumicolor, O. oenanthoides, and the two 
species of the leucophrys superspecies: O. 
leucophnjs and O. piurae) are character- 
ized by their being somewhat larger sized 
than the species of the other groups, by 
their relatively long and thin bill, and their 
brown or gray coloration. O. fumicolor is 
almost entirely dark rufous, O. oenantlioi- 
des is gray above and pale rufous below, 
whereas birds of the leucophrys super- 
species are gray both above and below. 
The birds of this species-group live higher, 
as a group, than any other in the genus. 
They are found in wet paramos (O. fumi- 
color) or in dry scrub associations of the 
puna (O. oenanthoides and O. leucophrys) . 

The Genus Xo/m/'s 
Diagnosis 

A genus of 13 to 14 species of medium 
to large sized flycatchers distributed along 
the high Andes, in the uplands and low- 
lands of central South America, and in 



194 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



\ 



\ 



LENGTH of TARSUS 
mm 

33- 
32- 

31- \^ 

30- ^\ 

29- 

28- 

27- o 

26- o ^ 

25- o qip 

24- •^ 

23- 



\ 



\ 



Q 


A 




A 


'\ D D 




\ 




\ 




o \^ 




^ \ 




CD \ 




O N 


s 



\ 



\ 



\ 



\ 



o X.striaticollis 

^ X.pernix 

• X.fumigata 

♦ X.fuscorufa 
° X.rufipennis 

^ X.erythropygia 



8 



-r 
9 



10 



-T" 

n 



LENGTH of CLAW of HALLUX 



12mm 



Fig. 6. Length of tarsus plotted against length of claw of hallux in the Xolwis lumigata species-group (1) and the X. 
erythropyg/o species-group (2). Note that tarsus and claw of hallux are proportionately longer in the two species of the 
erytfiropyg/a group, and are separated from the measurements of the fumigafo group. 



Patagonia, where they inhabit forest edges, 
woodlands, dense scrub, and, more rarely, 
open steppes. 

All species of Xolmis, except X. pyrope, 
are patterned, but the details of color and 
color combinations vary widely from spe- 
cies to species. The colors making up the 
patterns are white, gray, and black in some 
species (in the c'merea species-group), gray 
and brown in others (the fumigata, 
erijthropygio, and ruljetro species-groups). 
The underwing patterns are relatively uni- 
form throughout the genus (Fig. 2), al- 
though the colors can be either rufous and 
dark brown, or black and white, as was 
mentioned earlier. 

Species of the genus Xolmis have a long 
and relatively narrow bill with a terminal 
hook (see Fig. 4). There is interspecific 
variation in bill width within the fumigata 
species-group. All species of Xolmis, except 
the rather terrestrial species, X. rubetra. 



have moderately long and conspicuous 
rictal bristles. The following species have 
emarginated outer wing feathers: X. striati- 
collis, X. pyrope, X. dominicana, X. irupero, 
and X. rubetra (see Fig. 3). Only in the 
latter species is there any evidence of a 
sound being produced in flight, which may 
be attributed to this emargination (Wet- 
more, 1926a; see Smith). 

Species Included 

The genus Xolmis as understood here 
includes species formerly placed in the 
genera Myiotheretes Reichenbach, 1850; 
Cnemarchus Ridgway, 1905; Ochthocliaeta 
Cabanis and Heine, 1859; and Pyrope 
Cabanis and Heine, 1859. (One species, 
Xolmis mtirina, has been removed from 
Xolmis and transferred to Agriornis, and 
will be discussed below under that genus. ) 
The five above genera have been merged 
into one because neither morphological nor 



Evolution of Ground Tyrants • Smith and Vuilleumier 



195 



ecological characters permitted unequiv- 
ocal delimitation of more than one genus. 
Thus, X. pyrope cannot satisfactorily be 
separated from Xolmis sensu Hellmayr 
(1927), and X. rufipennis can just as easily 
be included in de Schauensee's (1966) 
enlarged genus Mijiothereies as in Xolmis. 
It is as arbitrary to shift either or both of 
these two species back and forth from one 
genus to another as it is to isolate them in 
monotypic genera, since neither species 
really departs from the species composing 
the genera Myiotheretes and Xolmis in de 
Schauensee's list. Another solution (merg- 
ing) is preferred here, but only if sub- 
divisions within the enlarged Xolmis can 
be expressed. This is done here by means 
of species-groups. 

1. The fumigata species- group. This 
species-group includes Xolmis striaticollis, 
X. pernix, X. fumigato, X. fusconifa, and 
perhaps also X. signata.^ 

These species have relatively broader bills 
than other species of Xolmis, a whitish or 
grayish-colored throat with dusky streaks, 
unpattemed smoke-brown or rusty breast 
and belly, conspicuously patterned wings 
and tail (a given species can have either 
the tail or the wings patterned, or both), 
and relatively short tarsi and claws of 
hallux. 

The species of the fumigata species- 
group resemble most those of the eryth- 
ropygia species-group in color and,^ pat- 
tern, but they differ from them in having 
proportionately broader bills (Fig. 5), 
shorter tarsi, and shorter claws of hallux 
(Fig. 6). These and other differences in 
proportions are illustrated in Figure 7 by 
means of a ratio-diagram. 



^ X. signata ( Taczanowski, 1874) is known from 
two specimens (fide Hellmayr, 1927). One of 
these is now deposited at tlie Polish Academy of 
Sciences in Warsaw ( Pinowski, in lift. ) . I have 
not been able to examine this specimen and can- 
not, tlierefore, make any decision al:)Out its taxo- 
nomic allocation. I retain X. signata in the fumi- 
gata species-group and, indeed, in Xolmis, imtil 
the specimen(s) can be critically studied. 



The habitats favored by the species of 
the fumigata species-group include cool, 
moist, forested or wooded slopes of the 
tropical Andes in the ecotone between the 
montane forests (cloud forests) and the 
grasslands above. They do not occur in 
high altitude treeless steppes, where Agri- 
ornis and Muscisaxicola live. They also 
seem to avoid the dark montane forest 
interior, where Ochthoeca cinnamomeiven- 
tris is found. 

The birds of the fumigata species-group 
converge both moq^hologically and in some 
of their behavioral traits toward thrushes 
of the genus Turdus. Note in particular the 
resemblance between X. striaticollis and 
Turdus spp. in color pattern and in stance 
when on the ground. 

2. The erythropygia species- group. This 
species-group comprises two sympatric, 
probably not very closely related species, 
Xolmis erythropygia and X. rufipennis. 
They differ, as a group, from the species 
of the fumigata species-group, which also 
have streaks on the throat, by the combi- 
nation of: relatively narrower bill, re- 
latively longer tarsus, and longer claw of 
hallux (see Figs. 5-7). They differ from 
the other species-groups of Xolmis in hav- 
ing conspicuously patterned underwings 
with rufous instead of white (see Fig. 2). 

X. erythropygia occurs in wet habitats at 
the upper altitudinal limit of montane 
Andean forests. X. rufipennis occurs in 
slightly drier and more open habitats, often 
within the puna grassland and scrub in 
Polylepis woodlands (Vuilleumier, 1969c). 
Both species, but X. erythropygia espe- 
cially, have some superficial resemblance 
to chats, rather than to thrushes. 

3. The pyrope species-group. The only 
species included, Xolmis pyrope, is gray 
above, with a whitish throat and belly and 
pale gray breast, without wing or tail 
markings, thus contrasting, in this lack of 
pattern, with all other species of the genus. 
Its red iris makes it resemble X. cinerea in 
this character, whereas its juvenal plumage 
resembles that of Muscisaxicola species in 



196 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



-15 

H 



-.10 

^-h- 



-.05 







.05 



.10 



WING <^.^ \ ^ 

TAIL '^''? 

W 

TARSUS ^ 

CULMEN <^<y^ Q< <s ^ 



(S 



\ 



\ 



---X.erythropygia 
••— X. rufipennis 

^-X.fumlgata 

X.fuscorufa 

•-"— X.pernix 






■■■"" X.striaticollis : 
STANDARD of COMPARISON 



Fig. 7. Ratio-diagram illustroting differences in proportions between members of the Xo/m/s lumigata and X. eryfhropyg/'a 
species-groups. Note that X. erythropyg/o and X. rufipennis (of the erythropygia group) have proportionately longer tarsi 
and shorter culmens than the species of the fumigata group. The scale is in units of difference from log of standard of 
comparison. For further information on the construction of ratio-diagrams, see Simpson (1941). 



having faint streaks on the breast and buffy 
coloration along the edges of wing coverts. 
X. ptjrope has markedly shorter tarsi than 
species of the cinerea and rubetra species- 
groups, and is thus closer in this character 
to X. fumigata of the fumigata species- 
group (see Fig. 8). 

X. pyrope lives in Patagonia along the 
edge of, and in open situations within, 
Nothofagus (beech) forests. 

4. The cinerea species-group. This group 
contains Xohnis cinerea, X. velata, X. coro- 
nata, X. dominicana, and X. irupero. These 
species are patterned in various combi- 
nations of white, gray, and black. X. 
cinerea has white throat and crissum and 
gray breast; the other four species have 
wholly white undeiparts. The details of 
patterns of wings and upperparts vary from 
one species to another. The possible rea- 
sons for the variation in color patterns have 
not been investigated. Friedmann (1927: 
197) believed X. irupero' s habit of nesting 
in holes in trees to be related to its black 



and white pattern, but he did not suggest 
how. In a general way, the species of this 
group have color patterns reminiscent of 
those of the genus Oenanthe. Cott (1947) 
showed that certain species of Oenantlw 
are unpalatable, and that their highly con- 
spicuous patterns may be inteipreted as 
warning coloration. It would be interesting 
to carry out experiments to check whether 
the same might be true in Xolmis also. In 
proportions, the five species of the cinerea 
species-group are rather similar to one 
another, except that X. dominicana has a 
longer tail (see Fig. 9). 

Birds of the cinerea species-group are 
arboreal and live in woodlands, savannas, 
and other nonforest vegetation types of 
central South America, from central Brazil 
southward to eastern Bolivia, Paraguay, 
Uruguay, and central Argentina. 

5. The rubetra .species-group. Xolmis 
ruhetra, the only species of this group, 
differs from the species of the cinerea 
species-group in having brown upperparts, 



Evolution of Ground Tyrants • Smith and Vuilleumier 



197 



-.15 

h 

WING 



TAIL 



TARSUS 



CULMEN 



.10 



05 

-h-H 





-h 



.05 



■H 




X.rubetra 

— X.pyrope 
^^X.fumigata 



^^X.cinerea: 
STAN DARD of COMPARISON 



Fig. 8. Ratio-diagram illustrating differences in proportions between selected species of tfie genus Xo/m/'s. Note that 
Xo/mis pyrope (sole member of the pyrope species-group) resembles more closely X. lumigata (representative of the 
fumigata group) than X. cinerea (representative of the cinereo group), Vi'hile the more terrestrial species X. rubetra (sole 
member of the rubetra group) differs from all of them in proportions of wing, tail, tarsus, and culmen. 



a collar of blackish streaks on the breast, 
rufous flanks, and proportionately long 
tarsi (compare X. cinerea and other species 
in Figs. 8 and 9). X. rubetra lives in open 
scrub habitats of central and southern 
Argentina, and appears more terrestrial 
than the other species of the genus. In 
color, pattern, and habits, X. rubetra is 
intermediate between the other species of 
Xolmis and Neoxohnis rufiventris, and may 
represent an evolutionary transitional 
"stage" between arboreal and terrestrial 
tyrants. 

The Genus Neoxolmis 

Diagnosis 

A monotypic genus of large terrestrial 
flycatchers living in open steppes of south- 
ern Patagonia. This genus is characterized 
by: gray throat and breast, chocolate 
brown belly and crissum ( recalling the pat- 
tern of Xolmis erythropygia), brown and 
rufous underwing pattern (Fig. 2), and 
conspicuous pale edges to secondaries and 



wing coverts (recalling the patterns of 
Xolmis velata and X. rubetra). The pro- 
portions of Neoxohnis differ from those of 
species of Xolmis; N. rufiventris has pro- 
portionately long wings and tarsi but short 
tail and bill (Fig. 9). The lores and 
adjacent parts of the face are dark sooty 
gray, giving the head an aspect reminiscent 
of that of Muscisaxicola macloviana, which, 
incidentally, also lives in southern Pata- 
gonia. Neoxolmis has veiy short and in- 
conspicuous rictal bristles. 

Some of the above characters may be 
correlated with the terrestrial behavior 
and nesting habits of this species (Craw- 
shay, 1907; Maclean, 1969): long legs, 
long wings, short rictal bristles. 

Hellmayr (1927: 39) erected the genus 
Neoxolmis for the species rufiventris, 
which had usually been placed in Myio- 
theretes. Zimmer (MS) placed it in Xolmis, 
but de Schauensee (1966) maintained 
Neoxolmis. 

From this history of taxonomic shifts, 
and from the intermediacy in characters 



198 Bulletin Miiscuin of Comparative Zoology, Vol. 141, No. 5 



'-^5 



WING 



TAIL 



TARSUS 



00 

-f- 



\ 



\ 



/ 



CULMEN d 



-.05 







.05 



.10 



.15 

■H 




•X.irupero 

'X.velata 

X.dominicana 

•X.coronata 

Neoxolmis 
ruflventris 



■""X.cinerea: 
STANDARD of COMPARISON 



Fig. 9. Ratio-diagram illustrating the relative proportions of the five species of the Xo/m/'s cinereo species group and of 
Neoxolmis rufiventris. X. dom'micana differs markedly from all species of the cinereo group in tail-length, and Neoxo/mis 
differs from all species of the c/nereo group in its proportionately much longer wings and tarsi, but shorter culmen. 



listed above, it is evident that Neoxolmis 
rufiventris is a problem species. In such 
cases, two alternatives occur for the taxon- 
omist: either to place the aberrant species 
in a monotypic genus, or to lump it with 
another genus. Neither of these decisions 
is devoid of subjectiveness. My reason for 
maintaining Neoxolmis rufiventris in a 
monotypic genus distinct from an enlarged 
genus Xolmis (to which it is most closely 
related) is that N. rufiventris has gone 
farther toward the terrestrial adaptive 
zone (zone sensu Simpson, 1953) than any 
other Xolmis, including X. ruhetra. 

The Genus Agriornis 

Diagnosis 

A genus of five species of medium- to 
large-sized flycatchers distributed along 
the Andes and in Patagonia in open, non- 
forest vegetation. The plumage of all spe- 
cies is dull grayish or brownish above, with 
or without obsolescent streaks, and grayish 
brown below with a buffy wash on the 



flanks and undertail coverts. The throat 
and chin are whitish and streaked with 
dark brown or blackish, conspicuously so 
in some forms ( see Fig. 1 ) . The two outer- 
most primaries are emarginated in adult 
males ( see Fig. 3 ) , but not in adult females 
or immaturcs of either sex. (Hellmayr's 
comments [1927] and footnotes about the 
emargination of outermost primaries in 
livida and microptera arc not accurate.) 
The function of this emargination is not 
known; its presence in adult males suggests 
that it might have something to do with 
some sort of display, but I have not found 
any support for this in the available litera- 
ture. I have seen A. alhicauda perform a 
flight display, but unfortunately the dis- 
playing bird could not be collected, and its 
sex is unknown. 

The remiges are not patterned, although 
the secondaries have buffy edges; the 
rectrices have either the outer web of the 
outermost rectrix whitish, or the outermost 
rectrices entirely or almost entirely white. 
The latter tail pattern is found only in 



Evolution of Ground Tyrants • Smith and Viiilleumier 199 



alhicauda and montono, and is geographi- 
cally variable in montana. 

The bill is strongest and most markedly 
hooked in the two largest species, alhi- 
cauda and livida, which feed, perhaps 
regularly, on small vertebrates (see Fig. 
4). Two other species (montana and 
microptera) also feed, at least occasionally, 
on small vertebrates, but miirina does not, 
and to my knowledge, is strictly insec- 
tivorous. Because of their aspect (bill 
especially) and habits (hunting), the 
largest species of Agriornis have frequently 
been compared to shrikes (Laniidae), a 
group absent from South America. In other 
ways, however, Agriornis flycatchers look 
and act more like thrushes than shrikes. 

Species Included 

Until now (Hellmayr, 1927; de Schauen- 
see, 1966), the genus Agriornis was 
considered to consist of four species: A. 
montana, A. alhicauda, A. livida, and A. 
microptera. In this paper, I suggest re- 
moving murina from Xohnis and placing it 
into Agriornis. (De Schauensee, 1966: 334, 
remarked that "in plumage pattern this spe- 
cies [murina] suggests a small Agriornis.") 

On the basis of absolute size, murina is 
smaller than any Agriornis, indeed it is 
smaller than any species of Xohnis except 
X. irupero. However, certain other char- 
acters exhibited by murina are, I believe, 
more conservative than size and are taxo- 
nomically more important above the" spe- 
cies level. 

There is much variation in proportions 
from species to species in Xohnis and less 
in Agriornis; it is obvious that murina re- 
sembles Agriornis more than Xohnis. In 
particular, the proportions of the small 
murina and the large A. alhicauda are 
rather similar (see Fig. 10). 

The general coloration and pattern of 
murina are not found in Xohnis, but are 
characteristic of Agriornis (see also remark 
in de Schauensee, 1966). No Xohnis is pale 
grayish brown and has a whitish throat 
with dark streaks, but these characters are 



observed in all Agriornis. The juvenal 
plumage of murina (not seen by me but 
described by Wetmore, 1926b: 447) ap- 
pears to be like that of Agriornis, especially 
A. microptera. 

Other characters, such as emargination 
of the two outermost primaries, general 
habits and behavior, and habitat, cannot 
be used because they are shared by Xohnis 
and Agriornis and are not good group char- 
acters. Nevertheless, it would seem that 
murina is more of a ground bird than are 
most Xohnis, thus resembling Agriornis, 
but X. ruhetra, at least, is just as much a 
ground tyrant and an undoubted Xohnis. 

In conclusion, murina seems unlike any 
Xolmis but is more like Agriornis in color, 
pattern, and proportions and should be 
transferred from Xohnis to Agriornis. 

1. TJie montana species- group. The two 
species of this group, A. montana and A. 
alhicauda, are very similar in color pattern 
of the tail, but differ in absolute size and 
in the color of the bill, especially the lower 
mandible (black in montana, horn-colored 
in alhicauda). They are widely sympatric 
in the high Andes; I have found the two 
species together in the same habitats in 
Bolivia (rocky slopes with boulders and 
screes, vegetation of grasses, and sparse, 
low shrubs), but montana was more com- 
mon than alhicauda. 

2. The livida species-group. A. livida and 
A. microptera are the two species of this 
group. They resemble each other in pat- 
tern but differ in size. A. livida is more 
rusty on the underparts than A. microptera. 
They are entirely allopatric, but both live 
in denser habitats than birds of the pre- 
vious group (see Johnson, 1967), at least 
in southem South America. The high 
Andean populations of microptera occur in 
very arid steppes (personal observ^ation ) . 
Both livida and microptera build their nests 
in shiTibs or trees. 

3. The murina species- group. The only 
species included is Agriornis murina, pre- 
viously included in the genus Xolmis. A. 
murina is much smaller than the other spe- 



200 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



"15 -10 -.05 .05 .10 

I I I I I I I I I I I I I I I I I I I I I I M I I 



WING 



\ 



TAIL 



\ 

TARSUS / 

• 

/ 

/ 

CUL- / 
MEN^ 




— "A. murina 

A.microptera 

•^— A.montana 

(NW Argentina) 

••- -A.albicauda 
- — A. livida 



I 
I 

6 6 

STAN DARD of COMPARISON 



'A.montana 
(Ecuador): 



Fig. 10. Ratio-diagram illustrating the relative proportions of the five species of the genus Agriornis. Agr/'orn/s murina, 
although smallest in absolute measurements, is similar in proportions to the four other species. Two geographically distant 
populations of A. montono (from NW Argentina and Ecuador) are included to show geographical variation in proportions. 



cies of Agriornis, and is not closely related 
to any of the four other species in the 
genus. A. murina lives in the open steppes 
of Patagonia. 

The Genus Musc/'sox/co/o 
Diagr\os\s 

A genus of twelve to thirteen species of 
small- to medium-sized flycatchers char- 
acterized by pale gray or sandy brown 
plumage, unpatterned except for a species- 
specific crown spot (Fig. 1) that either 
varies from vellow to ochre and chestnut or 
is absent, and a whitish or buffy outer edge 
to the outermost pair of rectrices. In fresh 
plumage, the tip of the tail is edged with 
pale gray, whitish, or pale buff, but in 
worn plumage, this edge may be altogether 
lacking. 

Muscisaxicola (including MuscigroUa) is 
a rather well-defined genus, with no close 
affinities with other tvrannid genera. 
Xolmis pijrope, although arboreal, actually 
resembles Muscisaxicola in having a similar 



Juvenal plumage, in lacking a color pattern 
(it is uniformly gray), and possibly also in 
vocalizations (Smith). The monotypic 
Lessonia rufa (placed by Hellmayr, 1927, 
immediately after Muscisaxicola) , although 
strikingly dimoiphic, has some resemblance 
to Muscisaxicola in mannerisms, bill shape, 
and female plumage, but this could be 
convergence, since Lessonia, like Muscisaxi- 
cola, lives in open habitats. (Ochthornis, a 
monotypic genus found along rivers over 
a large area of tropical South America, 
looks much like Muscisaxicola fluviatilis 
in some ways, but I suspect its relationships 
to be rather with the Ochthoeca group. ) 

Species Included 

Hellmayr (1927) included the following 
species in the genus Muscisaxicola (the 
sequence is his): rufivertex, alhilora, junin- 
ensis, flavimicha, capistrata, frontalis, 
alhifrons, alpina, cinerea, viacloviana, 
moculirostris, and fluviatilis. Later, in his 
work on the avifauna of Chile (1932), 



Evolution of Ground Tyrants • Smith and Vuilleumier 201 



Table 1. Comparison of characters of Muscwralla and Muscisaxicola 



Character 



Miiscigralla 



Muscisaxicola 



Relatixe tail-length In males, tail 62.6% of wing-length 

Relative tarsus-length Tarsus 24 to 28 mm 
General coloration Pale sandy-brown 



Crown spot 

Rump 

Uppertail coverts 
Tip of tail 



Yellow, more or less concealed 

Buffy-brown, in contrast with back 

Chestnut-brown 

Pale buff, especially in fresh plumage 



Egg number and color 3-5, usually 4; white 

Nest site On ground, protected by vegetation 

Nest shape Open cup in protected sites; % domed in 

more open sites 

Stance Upright 

Tail and wing motions Present 

Display flight and song Display flight tyrannid-like; vocalizations 

Muscisaxicola-MVe, but more complex 



In males, tail varies from 63.6% to 
76.5% of wing-length 

Tarsus varies from 19 mm to 40.5 mm 

Pale gray to buff-brown, including 
sandy-brown 

Ochraceous to yellowish and chestnut, 
not concealed 

Concolor with back 

Black 

Pale whitish buff, especially in fresh 
plumage 

Usually 3; white with reddish spots 

On ground, or in cavity, protected by 
rocks, pebbles 

Open cup ( always in protected sites ) 

Upright 

Present 

Display flight more complex; vocaliza- 
tions less complex than Miiscigralla 



however, he considered cinerea to be a 
subspecies of alpino. With this modifi- 
cation, his classification and sequence have 
been used by ornithologists working on 
Andean birds ever since, for example, by 
Bond and de Schauensee (1942), Bond 
(1947), Olrog (1963) and others. In this 
paper, oJpina and cinerea are once more 
considered separate species taxonomically, 
but their very close relationship is indicated 
by placing them in a superspecies. Further- 
more, as I will explain later, I think that 
the species should be grouped in a slightly 
different way. Finally, I believe that the 
monotypic Miiscigralla hrevicaiida of the 
xeric savannas from southwestern Ecuador 
to northern Chile is closely related to 
Muscisaxicola. 

Subgenus Muscigralla 

Hellmayr's (1927) classification followed 
largely Berlepsch's earlier scheme ( 1907 ) . 
These authors listed Muscigralla brevi- 



cauda as a member of the subfamily Fluvi- 
colinae (which includes such genera as 
Agriornis., Xolmis, Myiothcretes, Muscisaxi- 
cola, Ochthoeca, Satjornis, Knipolegus, and 
Pyrocephalus) and placed it at the end of 
this subfamily, between the monotypic 
genera Ochflwrnis Sclater and Tumhezia 
Chapman. Hellmayr (1927) went as far 
as saying that Muscigralla was "a genus of 
doubtful affinity, possibly not belonging 
to this family" ( Tyrannidae ) . Koepcke 
( 1964 ) , however, in her guide to the birds 
of Lima, gave Muscigralla hrevicauda the 
same vernacular name (Dormilona) as 
Muscisaxicola, and indicated that this spe- 
cies has tail motions similar to those of 
Muscisaxicola. 

Table 1 lists the main characters of 
Muscigralla and Muscisaxicola. Muscigralla 
appears to differ from Muscisaxicola by 
its small size (see Fig. 11), proportionately 
long legs, partially concealed crown spot, 
buffy-brown rump, and chestnut brown 



202 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



Wl NG LENGTH 

mm 

170n 

160 

150i 

140 

130H 

120 

noH 

100 
90- 
80- 
70- 



o^ 



^ 






*M.brevicauda 
^M.fluviatilis 
oM.maculirostris 
■ M.macloviana 
►M.alpina superspecies 
□M.flavinucha 
▼M.albifrons 



TAIL LENGTH 



40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 mm 

Fig. 11. Wing-length plotted against tail-length in selected species of the genus Muscisoxico/a. Muscisoxico/a a/b/7rons, 
the largest species of the genus, appears as isolated from the other species, as is M. brevicauda, the smallest species. 



uppertail coverts. Muscigrolla resembles 
Muscisaxicola, however, in its general 
coloration, nest site, nest shape, stance, tail 
and wing motions, and display flight and 
vocalizations (see Smith). 

There is no doubt that Muscigrolla hrevi- 
caudo is more different from all species of 
Muscisaxicola than any of the species of 
Muscisaxicola are from each other. Yet, 
Muscigralla appears more closely related to 
Muscisaxicola than to any other taxon, and 
I suggest formalizing this affinity by 
placing Muscigralla lyrevicauda in the 
genus Muscisaxicola, but in a subgenus of 
its own. 

Subgenus Muscisaxicola 

1. The maculirostris species- group. The 
two species (maculirostris, fhwiatilis) that 
I include in this group differ from all other 
species of Muscisaxicola, except Inevi- 
cauda, by their small size. Muscisaxicola 
maculirostris and M. fluviatilis are basically 



brown in color, thus differing from most 
other species of the genus, which are gray 
or grayish brown. (Here again brevicauda 
is more like maculirostris and fluviatilis 
than like the other species of the genus. 
See Table 1). M. maculirostris and flu- 
viatilis both have the base of the lower 
mandible flesh or yellowish in life, and 
both have the edges of the wing coverts 
and secondaries varying from pale brown- 
ish buff to rich cinnamon (but many indi- 
cations in the literature mention or imply 
species-specific differences in lower man- 
dible and wing color: I wish to emphasize 
that they do not exist). 

The differences between the two species 
are as follows: 

(a) M. fluviatilis has a proportionately 
shorter tail and a shorter tarsus than 
macidirostris; 

(b) In M. maculirostris belly and under- 
tail coverts are concolor with the buffy 



Evolution of Ground Tyrants • Smith and Vuillewnier 203 



breast, whereas in fhwiatiUs, belly and 
undertail coverts are whitish and in con- 
trast with the buffy breast; 

(c) M. maculirostris is a species inhabit- 
ing xeric regions in foothills and plateaus 
of the high Andes and Patagonia, whereas 
fhiviatilis occurs along the banks or near 
the banks of tropical streams east of the 
Andes. 

2. The madoviana species-group. Musci- 
saxicola macloviana and M. capistrata, two 
southern South American species, are very 
close to each other in size and proportions 
and also in head pattern. Both have dark 
foreheads and sooty brown loral regions, 
which give them a distinctive "dark- 
headed" aspect. M. capistrata has a dark 
chestnut bro^vn crown patch, whereas in 
M. macloviana this region is sooty brown. 

3. The riifivertex species-group. The 
three species I include in this group occur 
in the high Andes and are veiy similar to 
one another in size, proportions, and color. 
They are medium-sized, proportionately 
long-winged, and pale gray or grayish 
brown with a black tail and a reddish 
brown to pale buffy-brown crown patch. 

The alhilora superspecies consists of two 
allopatric species or semispecies, juninensis 
(Peru, Bolivia, northern Chile) and alhi- 
lora (Chile), whereas the third species of 
this group, M. rufivertex (Peru south to 
central Chile), is sympatric with them. 

4. TJie alpina species-group. This spe- 
cies-group includes a single superspecies, 
ranging widely in the high Andes from 
Colombia to Chile and Argentina. The 
birds are of medium size, are brownish 
gray to pale gray in color, have the outer 
web of the outermost tail feathers narrowly 
edged with whitish or buffy-\\'hite, and 
lack a contrastingly colored crown patch. 

5. The alhifrons species- group. The three 
species of this group are the largest of the 
genus. One, M. frontalis, has a distinctive 
head pattern with sooty brown forehead 
and a sooty brown line in the center of 
the cro^vn, contrasting with a white loral 



spot. This species also has a proportion- 
ately long bill. M. alhifrons and M. flavi- 
nucha are very similar to one another in 
general coloration, but differ in size 
(especially in wing length) and in crown 
color. Both alhifrons and flavinucha have 
a whitish loral spot extending into a super- 
cilium (visible in the field). 

PART II: VARIATION AND SPECIATION 
IN THE GENERA XOLMIS, NEOXOLMIS, 
AGRIORNIS, AND MUSOSAXICOLA 

In the first part of this paper, the fly- 
catchers placed in the genera Agriornis, 
Neoxohnis, Xohnis, Pijrope, Myiotheretes, 
Muscisaxicola, MuscigraUa, and Ochthoeca 
by de Schauensee (1966) were arranged 
into genera with limits slightly at variance 
with those of earlier classifications (in- 
cluding de Schauensee's). Furthermore, 
the taxa included in these modified genera 
were grouped into subgenera (in one in- 
stance), species-groups, and superspecies, 
in order to express the hierarchy between 
the categories genus and species in a 
manner not conveyed by earlier authors 
having worked with these flycatchers. 

In Part II, I will analyze geographical 
variation and species formation in these 
birds. The genus Ochthoeca was not in- 
cluded, however, because my study of 
variation at the species level in that genus 
was not as thorough as in the other genera. 
Each species of the genera Xohnis, 
Neoxolmis, Agriornis, and Muscisaxicola 
(which are all diagnosed in Part I) was 
investigated to detect what sort of popu- 
lation structure {sensu Mayr, 1959) it 
exhibits: presence or absence of geographic 
variation, presence or absence of geo- 
graphic isolates, degree of morphological 
differentiation of isolates when present, 
and instances of secondary contact involv- 
ing, or not involving, range overlap and 
hybridization. 

The characters selected to assess such 
population structure were: wing-length 
(chord), tail-length (from insertion of 
central pair of rectrices to tip of longest 



204 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



rectrix), culmen-length (from insertion of 
ramphotheca into skull to tip), tarsus- 
length (from the distal edge of the distal- 
most entire tarsal scute preceding the first 
scute of the digits, diagonally to the articu- 
lation of tarsometatarsus and tibiotarsus ) , 
color intensity, and color pattern. Intra- 
and interpopulation variation in these char- 
acters was studied by comparing series of 
specimens of comparable age, sex, and 
wear from different localities representa- 
tive of the entire geographical range of 
each species being analyzed. Each locality 
consisted either of a single collecting spot 
(rarely, since series from such a spot were 
generally very small), or of a group of 
collecting spots located within relatively 
close distances of one another. Further 
details about the methods used need not 
be given here, because they are explained 
in the first paper of this series on Andean 
birds ( Wiilleumier, 1968). 

I must stress here that I have not tried 
to delineate infraspecific categories (sub- 
species) within the species of flycatchers 
treated in this paper. My chief reason for 
not doing so is my acceptance of the criti- 
cisms of Wilson and Brown (1953). Con- 
sequently, I have often ignored subspecies 
in my discussions of geographic variation 
within species. However, whenever I felt 
that it would be helpful to the reader, I 
have referred to subspecies if they were 
available. Thus, it might be convenient to 
refer to a small, geographically disjunct 
and morphologically differentiated popu- 
lation by its subspecific name, rather than 
by a geographical, and perhaps more 
cumbersome, term. On the other hand, I 
liave deliberately avoided using subspecies 
wlien describing geographic variation of a 
continuous, especially clinal, nature; it is 
precisely in such instances that the criti- 
cisms of Wilson and Brown (1953) take 
full force. 

The emphasis throughout my work on 
speciation, whether the birds studied were 
Tyrannidae, as in this paper, or other taxa, 
as Fumariidae ( Vuilleumier, 1968) or 



Coerebidae (Vuilleumier, 1969a), has been 
to determine how continuous, or on the 
contrary, how discontinuous, the distri- 
bution of given species was. I paid par- 
ticular attention to the presence of disjunct 
geographical variation, tiying to answer 
the question whether such variation re- 
flected the presence of an actual gap in 
the distribution of that sjiccies, and 
whether that gap had any significance in 
preventing gene flow between the isolated 
populations. I consequently made no spe- 
cial effort to examine large series of speci- 
mens for species having a small range and 
which would be likely to be geographically 
invariable, or for species having a broad 
range but which appear to be moqDhologi- 
cally and distributionally very uniform over 
that range. Decisions about such differ- 
ences in population structure were made 
on the basis of all available evidence: 
museum specimens, literature records, and 
field work. I make no apologies if I have 
overlooked geographical variation where it 
may indeed exist but may only be detect- 
able to a worker using more refined tech- 
niques or many more specimens than I did. 
The following accounts are based on the 
examination of about 695 specimens 
(Xohiiis, about 150; Neoxohnis, 10; Aiirior- 
nis, 185; and Musci.saxicoki, 350) and on 
personal field work with 4 species of 
Xohnis, 4 of Ag^riornis, and eveiy species 
of Mtiscisoxicolo. 

The following abbreviations are used: 
AMNH = American Museum of Natural 
History, New York; MCZ = Museum of 
Comparative Zoology. 

The Genus XoJmis 

7. The fumigafa Species-Group 

Xolmis striaficollis 

Found mostly solitary or by twos 
(Koepckc, 1958: 168; personal obser- 
vation), X. striaticoUis occurs along the 
Andes from Venezuela and Colombia in 
the north to northwestern Argentina in the 
south. It hunts in the open, frequently on 



Evolution of Ground Tyrants • Smitli and Vuilleumier 205 



the ground, but requires some sort of 
shrubbery or trees for shelter. It is thus 
encountered along the edge of montane 
forests or in bushes, between 1500 and 3500 
meters. In most regions, X. striaticoUis 
seems to avoid either very \\'et or very 
xeric habitats, but in Bolivia and north- 
western Argentina, I found it also in dry 
quebradas or in open valleys with a sparse 
growth of low xeric shrubs. 

Geographical variation in X. striaticoUis 
can be detected in size and color. Birds 
from southern Peru, Bolivia, and north- 
western Argentina are smaller in wing- 
length and paler in general coloration than 
those from Venezuela to central Peru. The 
change from relatively large and dark- 
birds to small and pale ones (with fewer 
and narrower throat streaks ) seems to occur 
rather abruptly in southern Peru, and might 
be considered either as a step in a cline of 
north to south geographical variation, or 
as an instance of secondaiy hybridization. 
I have not undertaken a special study of 
this problem, which is left open to further 
investigation. Furthermore, X. striaticoUis 
has an isolated population in the Santa 
Marta Range of northern Colombia, but 
these birds do not appear to be moipho- 
logically differentiated from birds from 
other Colombian localities. 

The fumigafa Superspecies 

XoJmis fumiiiata, pernix, and fuscorufa 
are three closely related, allopatric species 
(Fig. 12). X. pernix is isolated in the Santa 
Marta Range of northern Colombia, X. 
fumigata is widespread in the Andes from 
Venezuela to Peru but does not occur in 
the Santa Marta Mountains, and X. fus- 
corufa replaces fumigata from southern 
PeiTi to Bolivia. It appears that pernix and 
fuscorufa both have relatively narrow 
ranges, peripheral to the wide range of 
fumigata, and the question arises whether 
tliey have evolved as peripheral isolates 
from a -groio-fumigata-MVc ancestral stock. 

The evaluation of tlie relationships of 



these three forms is rendered somewhat 
difficult by two facts. The first is that the 
two peripheral populations {pernix and 
fuscorufa), although farthest apart ge- 
ographically, look more similar to each 
other than either does to the central 
populations {fumigata). Such a phenom- 
enon (polytopic taxa) is actually not so 
rare, and has been discussed by Mayr 
( 1942 ) , but it is not always easy to suggest 
an explanation for it. The second fact, 
discussed in the chapter by Smith, is that 
of his having observed in Ecuador (in 
other words, within the range of fumigata) 
a bird looking somewhat intermediate be- 
tween pernix and fuscorufa. With much 
needed further data still pending, especially 
the collection of specimens from this popu- 
lation, I conservatively base my treatment 
of the three forms on present specimen in- 
formation. 

Xolmis pernix. This localized species is 
known from seven specimens collected 
around the turn of the century (Todd and 
Carriker, 1922) and nine others collected 
more recently (Wetmore, Zusi, in litt.) in 
the Santa Marta Range of northern Co- 
lombia. From this paucity of specimens, 
it would appear that pernix is rare, al- 
though, to be sure, ornithologists have not 
visited the Santa Marta Mountains very 
frequently. Like X. fumigata, X. pernix 
seems to be a forest-edge species (Todd 
and Carriker, 1922). X. pernix is identical 
in size and proportions to X. fuscorufa and 
X. fumigata, except that it has a longer 
bill (see Fig. 7). In color pattern, pernix 
is much closer to the geographically dis- 
tant fuscorufa than to fumigata. 

Xolmis fumigata. Xolmis fumigata is the 
only ^^'idespread species or semispecies of 
the fumigata superspecies. It occurs along 
the Andes from Venezuela through Colom- 
bia and Ecuador, and as far south as 
central Peru (Fig. 12). The scanty data 
on habitat preferences suggests that fumi- 
gata inhabits edge situations at elevations 
from about 2200 to at least 3600 meters. 
Leveque (1964: 59) saw the species "on 



206 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



Cesar 
Depression 



Uppe^ 

Maranon 



Xolmis fumigata superspecies 
X. pernix 

X. fumigata 
I X.fuscorufa 





RTo Apurimac'^ \^'^^*S&h^ ' 

^^^^ Bolivia 



Fig. 12. Geographical distribution of the Xolmis fumigata superspecies. Xolmis pernix is isolated from the nearest popula- 
tion of X. fumigata by the Cesar Depression. The Venezuelan isolate of X. fumigata, isolated from Colombian populations by 
the Rio Torbes Valley, is clearly differentiated morphologically, whereas the Peruvian isolates, separated from one another 
by the Upper Maranon, are not, or are less well marked. X. tuscorufa does not seem to occur northwest of the Rio Apurimac. 



several occasions" around Quito, Ecuador, 
but in spite of numerous hours of field 
\\ ork in apparently suitable habitats around 
Quito and elsewhere along the Andes, I 
was never able to even glimpse it. 

Examination of series of specimens from 
the entire range of the species revealed 
that there is practically no geographical 
variation in size (as measured by wing- 
length), and only minor variation in color. 
The only discontinuous variation occurs 
from Venezuela to northeastern Colombia. 
Birds from the Andes of Tachira (west of 
the Rio Torbes Valley), of Merida, and of 
Trujillo can be considered as an isolate 
(see Fig. 12) (subspecifically separated as 
lugubm) distinguishable from other popu- 
lations of Venezuela and northern Colom- 



bia by having the crissum ochraceous 
instead of sooty and concolor with the rest 
of the underparts. I suppose that the 
transition from ochraceous to dark sooty 
brown takes place across the dry valley of 
the Rio Torbes (see Fig. 12), an important 
barrier for Andean species, as I pointed 
out elsewhere ( Vuilleumier, 1969a, 1969b). 
The birds from southern Ecuador and 
northern and central Peru appear to be 
geographical!)' isolated from one another 
by the valley of the upper Rio Maranon 
(see Fig. 12), and other rivers in the 
northern Peruvian Andes. From my study 
of a small series of specimens from the 
populations living on either side of the 
upper Rio Maraiion, I would conclude that 
morphological differentiation is not corre- 



Evolution of Ground Tyrants • Smitli and V iiilletimier 207 



lated \\'itli the apparent barriers. Yet, 
Chapman (1927: 4) assigned subspecific 
rank (cojomorcae) to the population hving 
along the Pacific slope of the Andes west 
of the upper Rio Maraiion. The problem 
of whether or not the Peruvian isolates of 
X. fiimigota are truly differentiated from 
each other, and from Ecuadorean popu- 
lations, remains to be solved by examining 
more skins. 

Xolmis fuscorufa. Xohnis fiiscorufa is 
relatively rare, as I judge from the small 
number of specimens (about ten in major 
U. S. and European museums), but its 
range is more extensive than that of pernix. 
It has been collected at several localities in 
southern Peru ( Paucartambo, Oconeque) 
and in Bolivia (Chulumani, Pucuyuni, 
Tilotilo, Incachaca, San Cristobal) (see 
Hellmayr, 1927; Bond, 1947; Bond and de 
Schauensee, 1942; and Niethammer, 1956). 
To my knowledge, no data on ecology are 
available in the literature. X. fuscorufo 
differs from pernix (and for that matter, 
from the other species in the species- 
group) in lacking the throat streaks. In 
color pattern, fuscorufa resembles striati- 
collis in having rufous on the tail feathers, 
although the exact pattern is not identical 
(by contrast, pernix and fumigata have 
only the outer web of the outermost rectrix 
rufous ) . 

Xolmis signata 

This species is apparently knowna only 
from the two specimens from central Peru 
mentioned in Taczanowski's description 
(1874: 532-533), one of which, at least, is 
at the Polish Academy of Sciences in War- 
saw (Jan Pinowski, in litt.). Not having 
seen it, I cannot decide whether the allo- 
cation of signata to this species-group (or 
indeed to Xohnis) is correct. As I judge 
from the description, signata would seem 
distinctly smaller than any other species of 
the fumigata species-group and somewhat 
intermediate in color between fumigata 
and erythropygia (the latter species be- 
longing to the next species-group). 



2. The erythropygia Species-Group 
Xolmis erythropygia 

Xohnis erythropygia occurs from about 
3000 to 3800 meters in wet montane forests 
or shrubbery at, or close to, the timber line. 
Leveque (1964: 59) saw one pair in Ecua- 
dor "in a zone of low bushes and wet 
meadows." It is relatively rare, but might 
be fairly common locallv (see, e.g., Peters 
and Griswold, 1943: '313-314); fifteen 
specimens from the Santa Marta Mountains 
are cited by Wetmore (1946), but the 
species was overlooked there during the 
early explorations of this range. 

X. erythropygia occurs in the Santa Marta 
Mountains of northern Colombia (sub- 
species orinoma), and ranges from southern 
Colombia to central Peru (erythropygia), 
and from southern Peru to Bolivia (bolivi- 
ana). Geographical variation in this species 
seems very slight. Even the isolated Santa 
Marta population does not seem well 
differentiated (as I judge from Wetmore's 
description, 1946: 5). Nevertheless, the 
pattern of geographic variation of erythro- 
pygia is interesting because it parallels the 
differentiation showed by the X. fumigata 
superspecies of the fumigata species-group. 
In other words, in erythropygia, as well as 
in the fumigata superspecies, the peripheral 
populations (Santa Marta isolate, and 
populations from southern Peru and Bo- 
livia) show morphological divergence from 
the central populations (Colombia except 
Santa Marta, Ecuador, northern and central 
Peru). In the fumigata superspecies, 
morphological differentiation is extensive, 
and species status has surely been reached 
by the three taxa, whereas in erythropygia, 
the peripheral populations are only weakly 
differentiated, and can only be considered 
to be subspecifically distinct. 

Xolmis rufipennis 

Xolmis rufipennis ranges from northern 
Peru to north-central Bolivia (Cocha- 
bamba), and seems relati\'ely uncommon 
(about 30 skins in major U. S. and Euro- 



208 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



pean museums ) . There does not seem to be 
any geographieal variation in this species. 
Notes about tlie habitat have been pub- 
hshed elsewhere ( Vuilleumier, 1969c). 

3. The pyrope Species-Group 

Xolmis pyrope, the only species of this 
group, is a common species inhabiting 
both slopes of the Andes up to about 1500 
meters, where it is found in open wood- 
lands and forest edges from central Chile 
southward to Patagonia and Tierra del 
Fuego. Migrants reach northward to Ata- 
cama during the southern hemisphere 
winter (Goodall et al., 1957: 151; Johnson, 
1967: 265). 

X. pyrope varies geographically on the 
continent, but this variation is veiy minor 
and probabh' entirely clinal. However, the 
birds from the Island of Chiloe, off south- 
ern Chile, are larger than those from the 
neighboring mainland ones, and were as- 
signed subspecific rank (fortis) by Philippi 
and Johnson (in Goodall et al., 1957: 152- 
154). 

4. The cinerea Species-Group 

Xolmis velata, X. coronata, and X. do- 
minicana do not exhibit any geographical 
\'ariation. X. cinerea varies geographically 
in a minor way only; Hellmayr (1927: 12) 
did not recognize the two subspecies which 
liad been previously described. Similarly, 
X. irupero shows only slight geographical 
\'aiiation: the two subspecies described 
w(>re not deemed taxonomically distinguish- 
able by Hellmayr ( 1927) and Pinto ( 1944). 

There does not appear to exist any 
potential for further speciation in the 
cinerea species-group. On the one hand, 
the five species of this group do not seem 
to have isolated and morphologically dif- 
ferentiated populations that could be con- 
sidered as incipient species. On the other 
hand, there is extensive sympatiy of several 
species over a large part of central South 
America (Brazil south of the Amazon, 
eastern Bolivia, Paraguay, Uruguay, north- 
ern and central Argentina). Sympatry, 



combined with the morphological distinct- 
ness of the five species, makes it im- 
possible to even attempt to reconstruct the 
past history of speciation in this species- 
group. Of the five species, X. velata, X. 
cinerea, and X. irtipero appear to be com- 
mon or relatively common, but X. do- 
minicana and possibly X. coronata may be 
scarce or even rare. Hudson (1920: 142) 
said of X. coronata: "Old gauchos have 
told me that fifty years ago they were 
abundant all over the pampas, but have 
disappeared wherever the giant grasses 
have been eaten down and have given 
place to a different vegetation." Detailed 
study of the ecological preferences of the 
species of this group should be undertaken 
to determine to what extent the moqiho- 
logical differences they show can be cor- 
related with habitat selection, predators, 
feeding habits, and nest sites. 

5. The rubetra Species-Group 

The only species of this group, Xolmis 
ruhetra, does not exhibit any geographical 
variation. Its range includes the lowlands 
east of the Andes in northern Patagonia 
(during the breeding season) and the 
Argentine chaco and western pampas (in 
the southern hemisphere winter) (Olrog, 
1963; Hudson, 1920: 148. however, thought 
ruhetra not to be migratory). 

The Genus Neoxo/m/s 

Neoxolmis rufiventris, the only species 
of this genus, breeds in southern Patagonia, 
where its nest has recently been found by 
Maclean (1969); it winters in the Argen- 
tine pampas ( Olrog, 1963 ) , as far north as 
Uruguay (Cuello and Gerzenstein, 1962: 
126). There is apparently no geographical 
variation in this species. 

The Genus Agriornis 

1. The monfana Species-Group 

Agriornis montana 

This wide-ranging species occurs in the 
paramos of Colombia and Ecuador, in the 



Evolution of Ground Tyrants • Smith and Vuilleuinier 209 




Fig. 13. Geographical distribution of Agriornis montana. Colombian and Ecuadorean populations [solitaria] are isolated 
from Peruvian ones (inso/ens) by the Northern Peruvian Low and Upper Maranon barriers (see text). The populations of 
southern Peru, northwestern Bolivia and extreme northern Chile (b) are intermediate in tail pattern between those to the 
north (a) and south (c). See text for further details. 



210 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



puna of the high Andes of Peru, Bohvia, 
and Argentina, and in open habitats along 
the Andes (in some locahties at relatively 
low elevations) all the way to southern 
Patagonia, exeept Tierra del Fuego (Fig. 
13). A. montana shows some geographical 
variation in size, but an assessment of this 
variation was not possible beeause many 
specimens in my samples had molting or 
heavih' worn tails or wings. There is geo- 
graphical variation in body color, tail pat- 
tern, and perhaps in vocalizations. 

Variation in body color appears to be 
extensive and gradual, but I cannot say 
\\'hether it is truly clinal or not. The 
samples at my disposal consisted of speci- 
mens that had been collected in widely 
different years, and I found that individuals 
taken prior to about 1940 were generally 
darker than those taken since that time. 
Mor(^ uniform samples would be necessary 
for meaningful intersample comparisons. 

Geographical variation in tail pattern 
can be summarized as follows. Samples 
from southern Colombia to south-central 
Peru (Cuzco area) (labelled "a" on Fig. 
13) have the four outermost tail feathers 
white, each rectrix having a white shaft. 
The fifth outermost feather has some gray 
but is still extensively white. In samples 
from southern Peru (Puno) the extent of 
white on outer tail feathers is diminished, 
so that the third outermost rectrix has gray 
on the inner web. In the La Paz Depart- 
ment of Bolivia the samples examined show 
much variation, and specimens collected at 
localities only a few kilometers apart may 
differ in their tail pattern. This situation 
was noticed by Hellmayr (1927: 5-6), who 
described the variable populations as a 
new subspecies, interjnedio, labelled "b" 
on Figure 13 (see also the comments of 
Niethammer, 1956: 103). 

In 1967-1968 I attempted to collect large 
series of these birds in northwestern 
Bolivia, but the low population density of 
A. montana made it impossible. Neverthe- 
less, among the birds I collected, two taken 
within 4 kilometers of each other in the 



Ballivian area of the altiplano southwest of 
La Paz city exemplify this individual vari- 
ation very clearly. One specimen (AMNH 
793198, collected 21 October 1967) has the 
two outermost rectrices wholly white, the 
third has some grav at the base of the outer 
\'ane and along the edge of the inner vane, 
and the fourth has white only in the distal 
half. The second bird (AMNH 793199, 
collected 22 October 1967) has some gray 
at the base of the inner vane of the outer- 
most rectrix, whereas the second and third 
tail feathers from the outermost have much 
gray at the base. In both specimens, the 
shaft of the outermost rectrix has at least 
some gray: at the base in AMNH 793198, 
and the entire proximal half in AMNH 
793199. 

However, other specimens I collected in 
the La Paz Department (such as AMNH 
793202, taken along the eastern front of 
the Andes between Viloco and Caxata) 
and in the Potosi and Chuquisaca Depart- 
ments have the same tail pattern as birds 
from farther south along the Andes in 
northwestern Argentina and in the Andes 
from northern Chile to Patagonia (labelled 
"c" on Fig. 13). In all specimens I have 
examined from tliese regions, the three 
outermost rectrices have a white outer vane 
and a white tip. The fourth and fifth tail 
feathers are only white-tipped. 

In short, geographical variation in tail 
pattern does not seem pronounced either 
north of central Peru or south of north- 
central Bolivia. The populations living in 
a relatively narrow zone in southern Peru 
and northwestern Bolivia, around the basin 
of Lake Titicaca (Fig. 13), are inter- 
mediate between two types of tail pattern. 
Several genera other than Aii,riornis have 
species that exhibit geographical xariation 
in the same area, or else have allopatric 
taxa at or immediately below the level of 
species coming in contact in this zone (see 
Vuilleumier, 1969b). 

Detailed analysis of larger samples than 
I have seen will be necessaiy before it 
is possible to tnteipret the significance 



Evolution of Ground Tyrants • Smith and Vuilleumier 211 



Table 2. Geographical variation in wing-, tail-, tarsus-, and culmen-length in 

AgRIORNIS ALBICAUDA 



Populations 
( Males only ) 


Wing-length 

(mm) 

Range (mean) 

(N) 


Tail-length 

( mm ) 

Range (mean) 

(N) 


Tarsus-length 

( mm ) 

Range (mean) 

(N) 


Culmen-length 

( mm ) 
Range ( mean ) 

(N) 


Ecuador 


137.0-156.0 (143.2) 

(5) 


115.0-132.0 (120.0) 
(5) 


41.0-44.0 (42.8) 
(5) 


31.0-33.5 (32.2) 

(5) 


Peru 


136.0-145.0 (140.3) 

(7) 


116.0-124.0(119.2) 

(7) 


39.0^2.0 ( 40.9 ) 

(7) 


31.5-33.0 (32.2) 
(7) 


Bolivia 


133.0-138.0(135.5) 

(2) 


111.0 

(1) 


39.5-40.5 (40.0) 

(2) 


.33.0-33.5 (33.2) 
(2) 


NW Argentina 


130.0* 

(1) 


101.0* 

(1) 


40.0 

(1) 


36.0 
(1) 


Note: The single 
these two ] 


specimen from NW Argentina has worn tips to both tail and wing 
measurements ( * in the table ) are probably lower than those of birds 


: feathers, so that 
in fresh plumage. 



of the geographical variation in A. 
montano. Provisionally, I would consider 
this variation to be a case of secondaiy 
hybridization taking place before repro- 
ductive isolation was achieved by the 
ex-isolates. That hybridization, rather than 
primary intergradation, is taking place is 
suggested by the individual variation ob- 
served in northwestern Bolivian popu- 
lations, which thus resemble hybrid swarms 
in other species. 

Geographical variation in voice may take 
place in A. montana. Birds from Ecuador 
(see Smith) and from Bolivia (Department 
Potosi, personal obsei^vation ) appear to 
have very similar vocalizations. On the 
other hand, birds from northern Argentina 
(Tucuman, see Smith) have slightly lower 
pitched and shorter calls. It would be 
most interesting to check, by means of 
additional recordings, whether such vari- 
ation is indeed geographical, and, if so, 
whether it parallels the variation in tail 
pattern just described, or, on the contrary, 
is independent of it. 

Agriornis albicauda 

This species is similar to montana in both 
color and pattern, including the tail pat- 
tern. A. albicauda differs from A. montana 
in having broader streaks on the throat, in 



being much heavier (the average of three 
albicauda being 82.3 grams, and of six 
montana being 60.7 grams), and in having 
a thicker, more powerfully hooked bill with 
pale honi-colored lower mandible (the bill 
of montana is entirely black). A. albicauda 
is the only species of the genus Agriornis 
living strictly in the high Andes, where it 
occurs from Ecuador in the north to north- 
ern Chile and northwestern Argentina in 
the south. 

There is some geographical variation, as 
pointed out by Zimmer (1937: 2-3). He 
stated that Ecuadorean birds are larger and 
darker than Peruvian ones. In order to 
check whether the variation in size might 
be discontinuous, I have measured speci- 
mens from Ecuador, Peiii, and Bolivia. As 
can be seen from Table 2, there is very 
little variation in culmen-length, and only 
a slight cline of decreasing size from north 
to south in wing-, tail-, and tarsus-length. 

Agriornis alJ)icauda and A. montana are 
too extensively sympatric for any inference 
to be made about their past history, assum- 
ing, as I do here, that they originated from 
the same ancestral stock. The Andes could 
have been invaded at two different times 
by an early stock, or else speciation could 
have taken place in situ in the Andes. The 
two species share the same habitats: open 



212 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 




p ^ A. microptera andecola 

"^ A. microptera microptera 
(winter range) 

I A. microptera microptera 
(breeding range) 

♦ A.livida 



Fig. 14. Geographical distribution of Agriornis livida and A. microptera. A. livida and A. microptera are fully al- 
lopatric. Wintering populations of A. microptera microptera overlap slightly with resident ones of A. microptera andecola 
in northwestern Argentina. 



Evolution of Ground Tyrants • Smith and Vuilleuinier 213 



slopes and valley floors of high Andean Agriornis microptera 

valleys, and areas having sparse and xeric Agriornis microptera has two series of 
vegetation of low shrubs, with scattered populations, in Patagonia and in the high 
rocks and boulders used as observation Andes respectively, separated from one 
posts. In Bolivia, I found that montona another by a hiatus of over 1000 kilometers 
outnumbered albicaiida about ten to one (Fig. 14). The Patagonian birds (sub- 
in such habitats. In northern Chile, how- species microptera) breed in open brushy 
ever, A. albicaiida was "more plentiful" steppes from Neuquen and Rio Negro in 
than A. montana (Johnson, 1967: 233). the north to Santa Cruz in the south; they 

do not occur on Chilean territory or on the 

2. The livida Species-Group island of Tierra del Fuego. This fact is 

Agriornis livida interesting, because it seems to suggest that 

... ..11 A. livida (which does occur on Tierra del 
The distribution of this species mcludes p^^ ^^ ^^^ ^ microptera are replacing 
the dry lowlands west of the Andes in ^^^^ ^^^^ ecologically. This supposition 
ChUe from Atacama (up to about 2000 ^^^^^^^ confinned when the distribution of 
meters altitude, Goodall et al., 1957: 141) ^^^ ^^ ^p^^,.^^ .^ examined in Patagonia 
southward to about 40-41 S. In this ^^^^^^^ ^^^^j^. ^ ^.^.^^^ .^ restricted to 
region of south-central Chile, the Andes ^^^ ^^^^^^ ^^^^^^ foothills, whereas 
are much lower and are less of a barrier ^^^^^^^^^^^ i, fo^^d in the lowland steppes 
than they are farther north, and con- ^^^^^^^ ^^^^ ^^^^^^^ ^^^ Atlantic Ocean, 
sequently, A. livida is found m the moun- ^j^^ Patagonian populations of A. mi- 
tains, especially in dry valleys lying m ram ^,^p^g,„ lea^p theij- breeding grounds 
shadows (the Lonquimay Valley, tor ^^^^^^ February and migrate northward 
example). South of about 41 S, A. livida ^^ ^^.^ nonbreeding quarters in northern 
does not occur west of the Andes because Argentina and southern Bolivia (see Fig. 
this zone is clothed by Nothofagus forests, ^^^ Nonbreeding Patagonian A. microp- 
which constitute unsuitable habitats; it oc- ^^^^ ^^^^ ^^^^^ collected in the following 
curs only along the dry eastern foothills of Argentine provinces : Mendoza (March), 
the Andes, mostly on Argentine territory g^|^^ (March, April, May), Tucuman 
(see Fig. 14). A. livida thus appears to ^^^^^ j^^^ j^j^^^ j^^^^ (^^y) Santiago 
avoid both very xeric and very wet habi- ^^j g^^^^^ ^j^jy^ ^^^ ^^.^^ (August) 
tats: its center of abundance, accordmg to (specimens examined personally), Cor- 
Hellmayr (1932) and Goodall et al. (1957), ^i^^^^ Buenos Aires, and Santa Fe (accord- 
is in central Chile, from Aconcagua to Bio- ^^^ ^^ Hellmayr 1927). 
Bio, and again in south-central Patagonia '^^le high ' Andean populations of A. 
in Chubut and Aysen. microptera (subspecies andecola) are sup- 
Geographical variation in A. livida is p^^^^i ^^ ^^ resident. Their geographical 
minor, and apparently involves only wing- ^.^^g^ includes northern Chile ^(Goodall 
length and tail-length (as pointed out by ^^ ^j 3^957. ;^43) northwestern Argentina, 
Hellmayr, 1932, and as confirmed by my BoIj^I^ a„^| probably southern Peru. In 
examination of series of specimens). The northern Chile, Johnson (1967: 232) found 
samples seen, however, are not sufficiently a. microptera in tola (Baccharis) steppes, 
large for one to conclude whether this where it was "fairly plentiful." On the 
variation is clinal or not. The populations Bolivian altiplano in 1967-1968, I saw 
living on Tierra del Fuego, although iso- microptera in similar habitats, but only 
lated geographically from other Patagonian twice during several months of field work, 
birds, seem not to be moq^hologically and can only conclude that the population 
differentiated. density of the species is low. 



214 Bulletin Museum of Comparative Zoology, Vol 141, No. 5 




m. maculirostns 



Fig. 15. Geographical distribution of Muscisaxicola maculirostris. The two peripheral isolates (niceforoi in Colombia and 
rufescens in Ecuador) are separated from each other by an ill-defined barrier (marked ?), whereas Ecuadorean and Peruvian 
populations ore isolated by the Northern Peruvian Low. 



Evolution of Ground Tyrants • Smith and Vuilleiimier 215 



The populations of A. microptera found 
in northwestern Argentina seem to be com- 
posed of migratory birds from Patagonia 
during part of the year, and of resident 
ones during the breeding season. The 
mixture of populations between March and 
July appears confusing. The only definitely 
breeding specimen I have seen so far 
is MCZ 99388, an adult male in worn 
plumage collected 12 December 1918 at 
Penon, 4000 meters, in Catamarca. Other 
specimens collected at lower elevations 
seem to be migrants: for example, those 
collected bv Steinbach at Cachi, 2500 
meters (Salta), in March and April 1905. 
Olrog (1949b: 153), however, collected a 
specimen in the Aconquija mountains 
(Tucuman) at an altitude of 2900 meters 
in May, and allocated this bird to the local, 
breeding subspecies andecola. This identi- 
fication suggests that the Andean popu- 
lations breed at very high altitudes during 
the rainy season between November and 
March, then move to lower elevations, 
where they encounter nonbreeding indi- 
viduals that have migrated from Patagonia. 
More data are needed to fully elucidate 
the breeding and nonbreeding ranges of 
Andean "residents" in northern Argentina. 

In spite of the wide geographical gap 
between breeding Patagonian and Andean 
populations of A. microptera, morpho- 
logical differentiation between the two is 
minor, and involves the color of streaks on 
the throat and thickness of the bill (Hell- 
mayr, 1927: 4). 

3. The murina Species-Group 

Agriornis murina, the only species of this 
group, breeds in xeric scrub in Argentine 
Patagonia (Rio Negro, Neuquen, and Chu- 
but). It migrates northward to the low- 
lands of northern Argentina, southern 
Bolivia, and western Paraguay during the 
nonbreeding season. There does not seem 
to be any geographical variation in A. 
murina. 



The Genus Muscisaxicola 
The Subgenus Muscigralla 

The similarities and differences between 
Muscisaxicola Inevicauda and the species 
of the subgenus Muscisaxicola and my 
reasons for considering them all congeneric 
were given in detail in Part 1. 

M. brevicauda occurs from southwestern 
Ecuador along the foothills of the Andes 
and the Pacific coastal plain of Peru to 
extreme southern Peru. The highest alti- 
tude at which the species has been col- 
lected is 1100 meters at Yamana (Loja), 
Ecuador (R. A. Paynter, Jr.). Marchant 
(1960: 372-373) found it "commonest in 
open country with or without scattered 
bushes," and often in "surprisingly thickly 
wooded places." My own observations in 
southern Ecuador and northern Peru agree 
with Marchant's. 

M. brevicauda shows no detectable 
geographic variation in color, but birds 
from southern and central Peru (Lima, 
Pisco, Nazca) are a little larger in wing- 
length and tarsus-length than those from 
the Guayas Province, southwestern Ecua- 
dor, and Piura, northern Peru. 

The Subgenus Muscisaxicola 
7. The maculirostris Species-Group 
Muscisaxicola maculirostris 

This species is the most widespread of 
the genus, ranging from the Bogota region 
of the Eastern Andes of Colombia south- 
ward along the Andes all the way to Tierra 
del Fuego (Fig. 15). It is found in a 
variety of habitats, which all have one 
thing in common: they are quite xeric, or 
even semi-desertic. Altitudinally maculi- 
rostris has a wide range, from about 1000 
to about 4000 meters (and occasionally 
above ) . 

As might be expected from this geo- 
graphical range, there is geographical 
variation in this species. Color xariation 
is not clinal, but sharply discontinuous. 
Ecuadorean birds (subspecies rufescens) 



216 Bulletin Museum of Comparative Zoologij, Vol. 141, No. 5 



altitude 
m 



4000- 



2500- 



1000- 



o 
o 
o o 



o 

o o 

o P 

o o 



o o 



•Ecuador and 
N. Peru 

oS. Peru and 

NW. Bolivia 

^N. Patagonia 



A A 



^ i A 



78 



— I — I — r 
85 



— 1 wing length 
91 mm 



Fig. 16. Altitudinal variation in wing-length in Muscisaxicola maculirostris. Andean populations from Ecuador southward 
to Bolivia show an increase positively correlated with altitude, but Potagonian birds do not. 



are very buffy, even pale rufous, on their 
underparts. In contrast, birds from north- 
cm Peru to Argentina and Chile are all 
very uniform (maculirostris), and of a 
much paler and more isabelline tone than 
Ecuadorean birds. This variation is shown 
in Figure 15. Interestingly, and perhaps 
unexpectedly, the birds from Colombia 
(niceforoi) do not resemble the geographi- 
cally close Ecuadorean birds, but are more 
like those from Peru to Argentina and 
Chile, much farther south. 

Geographical variation in size as mea- 
sured by wing-length is somewhat ob- 
scured by altitudinal variation (see Fig. 
16). Birds from southern Peru and north- 
western Bolivia live very high (3000 to 
4000 meters) and have long wings. Birds 
from northern Peru and Ecuador live at 
lower elevations ( 2000 to 3500 meters ) and 
have shorter wings. There is, however, no 
correlation between altitude and wing- 
length in birds from northern Patagonia. 
They live at low elevations (below 1500 
meters) but have long wings. 



Geographical variation in wing-length 
does not seem to parallel geographical 
variation in color: there is apparently no 
more abrupt change in wing-length be- 
tween southern Ecuador and northern Peru 
(see Fig. 16) than between other ap- 
parently not disjunct populations (e.g., 
Bolivia and northwestern Argentina). 

Muscisaxicola fluviatilis 

This is the only species of the genus to 
live in wet tropical lowlands and Andean 
foothills east of the Andes in Peru and 
Bolivia, and along streams and rivers in 
extreme western Brazil. However, there 
are in the literature a few reports of M. 
fhwiatilis at high altitudes in the Andes, 
which are worth examining in detail. 

Bond (1947: 130) synonymized M. titi- 
cacae (Carriker, 1932: 459) with M. fluvi- 
atilis. The type and only specimen of 
titicacae is an adult from Desaguadero 
along the shore of Lake Titicaca, at about 
3900 meters. I have seen this specimen and 
can confirm its identification as fluviatilis. 



WING LENGTH <^ 
90 



Mainland — Breeding 

RIO NEGRO(2) 
SOUTHERN CHILE (4) 

Mainland — Migrants 
CENTRAL CHILE(7) 

argentina(i) 

Falkland Islands- 
Breeding (3) 



Evolution of Ground Tyrants • Smith and Vuilleumier 217 



100 



110 11 5 



-f 



WING LENGTH ? 

90 



Mainland — Breeding 

RIO NEGRO(2) 
SOUTHERN CHILE(3) 

Mainland —Migrants 

CENTRAL CHILE a.lz) 
b.(6) 

argentina(i) t 

Falkland Islands-- 
Breeding(3) 



100 



110 115 



mm 



+ 



Fig. 17. Geographical variation in wing-length in mainland and Falkland Islands populations of Musc/saxico/o macloviana. 
In both moles and females, Falkland Islands birds have longer wings than do mainland birds, whether the latter are breed- 
ing or migrants. Number of specimens in parentheses. Females from Central Chile: a, with MCZ 94525, labeled " 9 ," but 
probably wrongly sexed; b, without this specimen. 



Olrog (1963) mentioned that fhwiatilis 
occurs in the highlands of northwestern 
Argentina. Through Olrog's courtesy, I 
examined four of the five specimens upon 
which this statement was based. Two of 
these five birds had been identified as 
M. fhwiatilis by Hellmayr (Olrog, in lift.). 
The four specimens I studied are un- 
doubtedly M. maculirostris, and I presume 
that the fifth belongs to this species also, 
since it is identical to them ( Olrog, in lift. ) . 

These specimens show not only that 
great caution must be exercised when 
identifying fluviatilis, but also that the 
species does occasionally occur high up 
in the Andes. 

There is apparently no geographical 
variation in M. fluviatilis. 

2. The macloviana Species-Group 

Muscisaxicola macloviana 

Muscisaxicola macloviana breeds in con- 
tinental southern South America, from 
about 41 °S, along the Andes southward to 
Tierra del Fuego, and in the Falkland 
Islands. Olrog (1948) noticed this species 
in flat country, along forests, and on sea- 
shores, as well as above the timber line in 
Tierra del Fuego. In the Falklands, Pet- 



tingill (personal communication) remarked 
that it was a "common resident in upland 
areas, usually in valleys near cliffs or rocky 
outcrops." He noticed further that "its 
particular habitat seemed to be governed 
by the availability of crevices in loose rock 
or ledges for nesting and grassy places, 
sheltered from the wind, for feeding and 
resting." 

The mainland and Tierra del Fuego 
birds (subspecies mentalis) are very uni- 
form in color and size, but have noticeably 
shorter wings than birds from the Falkland 
Islands {macloviana) (Fig. 17). I did not 
detect any parallel color variation. 

M. macloviana is migratory. The main- 
land populations move northward along 
the Andes in both foothills and uplands, 
and spend the southern hemisphere winter 
in Peru, where, according to Koepcke 
(1964), the species is regularly found in 
loose flocks in the coastal lomas. The Falk- 
land Islands populations, in contrast to the 
mainland ones, seem to be resident, al- 
though neither Bennett ( 1926) nor Cawkell 
and Hamilton (1961) say so explicitly. The 
resident status of Falkland Islands birds 
seems confirmed by an examination of 
wing-lengths among continental migrants, 
none of which were large enough to belong 



218 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



to the Falkland Islands populations (Fis;. 
17). 

Muscisaxicolo capistrata 

The breeding range of this species in- 
cludes extreme southern Chile and Tierra 
del Fuego, but very little is known about its 
breeding habits and ecological preferences. 
Neither Goodall et al. (1957) nor Olrog 
(194(S) give any data about its habitat. 

Muscisaxicoh capistrata migrates north- 
\\'ard along the Andes at high elevation, 
and spends the southern hemisphere winter 
in northern Chile, Bolivia, and southern 
Peru. There is no geographic variation in 
this species. 

3. The rufivertex Species-Group 
Muscisaxicolo rufivertex 

Muscisaxicola rufivertex breeds from 
northern Peru to central Chile and adjacent 
Argentina. In Chile (Goodall et al, 1957), 
this species vmdertakes altitudinal mi- 
grations and spends the nonbreeding season 
in the coastal regions of Coquimbo, Ata- 
cama, and Antofagasta. In Argentina, 
Boli\'ia, and Peru, I found M. rufivertex to 
inhabit xcric slopes with a rocky or gravelly 
soil and sparse shrubby vegetation, and 
rocky quebradas or small valleys with 
cliffs and rocky ledges. Dorst ( 1962 ) found 
rufivertex where small valleys open up 
into pampas, so that the ecological char- 
acteristics of grassy plains and rocky slopes 
of small valleys are combined. M. rufi- 
vertex has a rather extensive altitudinal 
distribution from about 3000 meters 
(Koepcke, 1964) up to about 4500 meters 
( personal observation ) . 

Muscisaxicola rufivertex has some geo- 
graphical variation in color and size. The 
detection of geographic variation is prob- 
ably complicated by migratory movements 
of several populations. Zimmer (1937: 3) 
stated that "one specimen from Cucha- 
cancha, Bolivia, belongs to palUdiceps 
although eleven other specimens from the 
same locality are occipitalis. The speci- 



mens were all taken in June, and there is 
little doubt that the individual belonging 
to pallid iceps was a migrant from the 
south." 

Variation seems to be as follows: north- 
ern Peruvian birds are the largest, whereas 
birds from Chile (Santiago region) are 
smaller. The birds from northern Chile, 
northern Argentina, and southern Bolivia 
are distinctly paler than those from central 
Chile and adjacent Argentina, or than 
birds from Peru and northern Bolivia. The 
palest birds live in northern Chile and 
northern Argentina. 

The albilora Superspecies 

Muscisaxicola juninensis. This species is 
found apparently only at very high alti- 
tudes from central Peru to Boli\ ia, northern 
Chile, and northwestern Argentina, where 
it seems to be resident. It inhabits grassy 
steppes interrupted by rocky outcrops and 
small cliffs, above 4000 meters, and usually 
between 4200 and 4S00 meters (personal 
observation ) . 

Muscisaxicola albilora. Muscisaxicola al- 
bilora breeds in the Andes of Chile and 
Argentina from about 30 °S to southern 
Patagonia, but not in Tierra del Fuego. It 
occurs at altitudes from about 1500 to 2500 
meters (Goodall et al., 1957). It seems to 
prefer rocky, barren slopes with almost no 
vegetation, at least in Rio Negro, Argentina 
( personal obsen^ation ) . 

There is no detectable geographical varia- 
tion in this species. Migrations are ex- 
tensive, since M. albilora vacates its 
breeding grounds (Johnson, 1967) and 
spends the southern hemisphere winter in 
the high tropical Andes, as far north as 
Ecuador. 

4. The alpina Species-Group 

The alpina Superspecies 

Hellmayr (1927) separated the various 
populations of Muscisaxicola having gray 
plumage and lacking a colored crown-patch 
into two species: Muscisaxicola alpina, 



Evolution of Ground Tyrants • Smith and Vuilleumier 219 





Muscisaxicola 
alpina superspecies 

back dark gray-brown 

back gray 

bock pale gray 

grisea (alpina 

semispecies) 

• ^ cinerea(cinerea 

semispecies) 



Fig. 18. Geographical distribution of the Muscisaxicola alpina superspecies. Dark-backed populations [alpina semispecies) 
have isolates in Colombia and Ecuador. In western Bolivia (and perhaps also in Peru: see text) the alpina semispecies comes 
in contact with pole-backed populations [cinerea semispecies). Insert: collecting localities, largely from personal field work, 
showing the contact area. Birds of the grisea phenotype [alpina superspecies) occur along the eastern front of the Andes. The 
cinerea phenotype [cinerea superspecies) occurs along the oltiplano as well as along the eastern Andean front. Localities 
marked with an asterisk (*) represent possible overlap. 



220 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



ranging from Colombia southward to Ecua- than the Andes of Peru-BoHvia, and espe- 
dor, Peru, and Bohvia in the department of eially those of northern Argentina and cen- 
La Paz, and M. cinerea, ranging from tral Chile (Lauer, 1952; Franze, 1927; 
northwestern Argentina to central Chile. Knoch, 1930). One might therefore con- 
In a footnote (1927: 28) he indicated that elude that geographical variation in color 
M. alpina is larger and darker, has whiter follows Gloger's rule. 

edges to secondaries and wing coverts, and Geographical variation in size, especially 

has a broader supraloral streak than M. as measured by wing-length, is extensive, 

cinerea. In a later publication, Hellmayr but does not follow exactly the same trend 

(1932: 125-126) merged the two species as variation in color. There is an increase 

into one, M. alpina. His main justification in size between Colombian and Ecua- 

for this taxonomic decision was that the dorean birds, the latter being the largest 

populations from northwestern Argentina of any populations of this group. From 

(Tucuman) were intermediate between Ecuador southward to Peru and Bolivia, 

M. alpina and M. cinerea: in size, these there is a decrease in mean wing-length 

Argentine birds were similar to M. cinerea, among the samples I have examined. This 

but in color they resembled M. alpina. decrease continues into northern Argentina 

Subsequent authors have followed Hell- and central Chile, so that Chilean birds 

mayr's 1932 treatment of alpina and cinerea are the smallest of any in the group. On 

as conspecific (see Olrog, 1963: 246; Phi- the whole, however, the north to south 

lippi, 1964: 141; Goodall et al., 1957: 164; decrease in wing-length cannot be said to 

Johnson, 1967: 244-245; de Schauensee, be clinal, because of some irregularities, 

1966: 336). which are worthy of careful examination. 

Geographical variation in the alpina- Indeed, within samples from central and 

cinerea complex is detectable in both color south-central Peru and Bolivia, variation 

and size, and can be described as follows, in wing-length is sharply bimodal, with 

There is a decrease in color saturation (of part of the birds in a sample having rela- 

both upper and underparts) from Colom- tively small wings and part having larger 

bia southward to Argentina and Chile, wings. Furthermore, the smaller birds also 

Samples from Colombia and Ecuador are appear to be paler than the larger ones, 

uniformly dark, especially on the upper- thus suggesting sympatry of differentiated 

parts, whereas samples from Peru and populations. 

Bolivia are distinctly paler, and finally, The published evidence bearing on this 
samples from northwestern Argentina and problem is as follows. In Huancavelica, 
central Chile are palest. Although this south-central Peru, Morrison ( 1939 ) col- 
variation is gradual, there is a clear-cut lected both large-dark and small-pale birds, 
break between Ecuadorean and Peruvian His six large-dark specimens (four males 
samples. This break corresponds to an area and two females) were collected between 
between the two countries where the Andes 12 and 25 October. As I judge from the 
are too low for birds of this group to occur, data on labels, the six males have relatively 
and where the dry to arid valleys, includ- large testes, but the females have small 
ing the Marafion valley, cut across the ovaries. All six specimens have slightly 
Andean ranges and are a barrier to gene \\'orn plumage. Taxonomically, the six 
exchange between birds living on either large-dark birds were assigned to the sub- 
side (Fig. 18). The geographical variation species grisea (locally breeding), and the 
just described appears to be positively cor- two small-pale birds to the subspecies 
related with rainfall. As a whole, the cinerea (migrants from Chile). It is note- 
Colombian and Ecuadorean Andes receive worthy that both taxa were collected in 
more rain, and have longer rainy seasons, October, at a time when local Andean 



Evolution of Ground Tyrants • Smith and Viiilleumier 221 



populations normally start breeding. Thus, 
either the two specimens of cinerea were 
late in beginning their southward migration 
to their breeding quarters in Chile, or else 
they were actually breeding in Peru. 

Bond (1947: 130), reporting on birds 
collected in Peru by Carriker, noted that 
four small-pale specimens, which he as- 
signed to the subspecies cinerea, "are adult, 
and were taken from April 10 to May 16." 
Since these dates correspond to the non- 
breeding season of many high Andean birds 
(Dorst, 1955, 1962; personal observ^ation ) , 
Bond concluded, I think correctly, that 
these individuals are "winter residents." 
These remarks, however, would not apply 
to the birds collected by Morrison (1939), 
cited previously. 

In Bolivia, Bond and de Schauensee 
(1942: 340-341) reported that griseo 
(large-dark) and cinerea (small-pale) had 
been taken together at three different 
localities: Viloca (La Paz), Cerro del Juno 
(Cochabamba), and Padilla (Chuciuisaca). 
A specimen of cinerea "was taken on the 
Cerro del Juno, October 2, where grisea 
was nesting at this time." This record is 
similar to the one published by Morrison 
(1939). In Bolivia, as in Huancavelica, it 
seems surprising that birds should remain 
as late as October in their winter range. 
Some individuals may stay longer than 
most wintering birds, as is well known 
from migrants in the northern hemisphere. 
Similarly, in the postbreeding season, some 
migrants are earlier arrivals than the bulk 
of the populations. Thus, the "three full- 
grown but immature specimens [of cinerea] 
. . . taken at Viloca [Department of La 
Paz], March 25 and 26" (Bond and de 
Schauensee, 1942: 341) may have been 
among the first wave of migrants. I believe 
that immature individuals of the genus 
Miiscisaxicola may be migrating earlier 
than adults. Thus, I have observed im- 
mature M. capistrata, a species breeding in 
southern Patagonia, in central Chile on 19 
March and in northwestern Argentina on 
25 April. By analogy, the arrival of im- 



mature birds from Chilean populations of 
cinerea in Bolivia in March would there- 
fore not be surprising. 

Two alternative explanations are possible 
for the above data. The first is that the 
small-pale populations ( taxonomically as- 
signed to Chilean cinerea) found in March, 
April, May, and October in Peru and 
Bolivia are migrants and do not breed in 
these last two countries. The second is 
that, even though the Chilean populations 
(cinerea) may be migratory, there are two 
moq^hologically differentiated populations 
of the alpina species-group living sym- 
patrically in parts of the Peruvian and 
Bolivian Andes. The second alternative 
implies that M. alpina (as understood, for 
example, by de Schauensee, 1966) consists, 
in fact, of two largely allopatric species 
having a narrow zone of overlap in the 
central Andes. 

Between October 1967 and January 
1968, I investigated the relationships of the 
populations of M. alpina in the Bolivian 
Andes. I traveled on the altiplano south- 
west of the Cordillera Real, and made 
several trips along the Cordillera Real, the 
Cordillera de Cochabamba, and, farther 
south still, in the Cordillera de Mochara. 
I visited these areas because they were 
located immediately around, as well as 
in, the region of possible sympatry of 
two differentiated populations of M. alpina. 
I wanted to determine in what habitats 
these birds occur, to collect samples of 
specimens in breeding condition, and to 
delimit the ranges of the two populations. 
Unfortunately, much more time was spent 
in just traveling large distances on ex- 
tremely difficult roads looking for these 
birds in habitats which appeared suitable, 
than in actually collecting and studying 
these birds. The extreme shyness of Miisci- 
saxicola, combined with their low popu- 
lation density even in the most suitable 
localities, was a powerful obstacle to col- 
lecting large comparati\'e series, which I 
had hoped to assemble. Furthermore, great 
physical difficulties were encountered in 



222 



Bulletin Museum of Comparative Zoologij, Vol 141, No. 5 



Table 3. A comparison of morphological characters in populations of the 

MUSCISAXICOLA ALPINA SUPERSPECIES FROM THE BOLIVIAN AnDES 



M. alpiua 
semispecies 



^f. cincrea 
semispecies 



Character 



Range (mean) 
(samijle size) 



Range ( mean ) 
( sample size ) 



Mensural characters 

Weight in giams (males and females) 24.9-27.9 (26.61) 

(7) 



Weight in grams ( males only ) 
Wing-length ( mm ) ( males only ) 
Tail-length (mm) (males only) 
Culmen-length ( mm ) ( males only ) 
Tarsus-length (mm) (males only) 



25.8-27.9 (26.92) 

(5) 

115.5-122.0(119.50) 

(5) 

78.5-90.5 (82.80) 
(5) 

17.5-19.0 (18.30) 

(5) 

32.0-33.5 (32.80) 

(5) 



Nonmensural characters 
Superciliaiy stripe 



Underparts 



Upperparts 
Wing coverts 



Whitish supercilium extends 
slightly beyond the eye 

Alidomen white, tinged with pale 
buffy 

Breast gray, darker than the 
whitish throat 

Uniform dark grayish l^rown 
( mouse-gray or mouse-l^rown ) 

Brownish gray, without buffy 
wash 



17.5-23.3 (20.55) 
(14) 

18.9-22.7 (21.06) 
(8) 

108.0-114.0 (111.87) 
(8) 

72.0-78.0 (75.18) 
(8) 

17.0-18.5 (17.93) 

(8) 

28.0-30.0 (29.37) 
(8) 



Whitish supercilium does not ex- 
tend lieyond the eye, at least in 
most specimens 

Alidomen whitish, tinged with pale 
gray or grayish brown 

Breast and throat pale gray, almost 
concolor 

Uniform grayish with distinct pale 
brownish tinge in most specimens 

Brownish with pale buffy wash 



the pursuit of these ekisivc birds at alti- 
tudes usually over 4000 meters. 

In spite of these difficulties, however, 
my efforts proved successful. I collected a 
total of 21 specimens of the alpina group, 
as well as other species of Muscisaxicola. 
This number is small, when compared to 
the yield of similar field endeavors in 
localities which are easier of access. Yet, 
the results of this exploratory collection 
permit the following unequivocal con- 
clusions. 

(A) Two populations of the olpina spe- 
cies-group occur in the high Andes of 
Bolivia (Departments La Paz and Cocha- 



bamba). One of them comprises large and 
relatively dark birds, hereafter referred to 
as "gr/.sert phenotype." The second con- 
tains small pale birds, referred to below as 
"cinerea phenotype." 

The birds of the grisea and cinerea 
phenotypes that I collected in Bolivia can 
be distinguished from each other by a 
series of characters (see Table 3). Weight 
alone permits complete separation, since 
there is no overlap in weight ranges be- 
tween the grisea and the cinerea specimens. 
In other mensural characters, birds of the 
grisea phenotype are consistently larger 
than those of the cinerea phenotype. In 



Evolution of Grou>jd Tyrants • Smith and Viiilleiimier 223 



color characters, birds of the griseo pheno- 
type appear generally darker than those of 
the cinereo phenotype. 

(B) Birds of both the grisea and the 
cinerea phenotypes breed between October 
and January. Although no nest could be 
located, the specimens collected include 
adult males with enlarged testes and fully 
ossified skulls for both phenotypes. The 
females collected did not have enlarged 
ovaries, however. No nestlings were col- 
lected, but two juvenile specimens of 
grisea had largely unossified skulls, either 
very small or unobserv^able gonads, and 
loose, very fresh plumage. All the other 
specimens, of both grisea and cinerea 
phenotypes, had worn plumage, especially 
remiges and rectrices. Food carrying and 
display behavior, including display flights, 
were seen from birds of the cinerea pheno- 
type in October, November, and Decem- 
ber. (The conditions of the gonads of the 
other three species collected, M. rufivertex, 
M. jiininensis, and M. alhifrons, were 
similar to those of the birds of tlie grisea 
and cinerea phenotypes.) 

(C) Birds of the cinerea phenotype oc- 
cur both on the altiplano and along the 
high mountains forming the eastern An- 
dean wall (Cordilleras Real and of Cocha- 
bamba) (see Fig. 18). Birds of the grisea 
phenotype, however, were encountered 
only in the Cordilleras Real and of Cocha- 
bamba (see Fig. 18). 

( D ) In one of the ten localities at which 
I collected specimens of the grisea or 
cinerea phenotypes, I observed both to- 
gether. Along the road to Morochata, at 
about 4040 meters altitude in the Cordillera 
Tunari ( Department Cochabamba ) , I en- 
countered a sparse population of Musci- 
scixicola on gentle, grassy slopes. The birds 
were foraging in the densely matted, 
grazed grass or turf, and were very wary. 
Much stalking and patience permitted me 
to secure five specimens within a radius 
of about one kilometer. 

Two of these birds proved to be of the 
cinerea phenotype. They had testes 9.0 X 



4.5 and 9.0 X 4.0 mm respectively, fully 
ossified skulls, and slightly worn plumage. 

The other three specimens were of the 
grisea phenotype. All three were males, 
with testes and skull conditions as follows. 
Testes 5.0 X 3.0 mm, skull with two tiny 
unossified windows; testes 6.0 X 3.0 mm, 
skull fully ossified; testes 4.0 X 2.5 mm, 
skull with a moderately large unossified 
occipital window. All three had slightly 
worn plumage, like that of the birds of the 
cinerea phenotype. 

The evidence presented above suggests 
that two allopatric species or semispecies 
come in contact along the eastern wall of 
the Andes of Bolivia (Department Cocha- 
bamba). An unequivocal decision on 
whether or not the taxa involved are spe- 
cifically distinct cannot be made from the 
present, still insufficient, data. More in- 
formation is needed on the respective distri- 
butional ranges of the grisea and cinerea 
phenotypes, on possible differences in 
ecological preferences, in displays or other 
behavior, and on the relationships in nar- 
row zones of overlap, as in the Cordillera 
of Cochabamba. Do the two taxa hybrid- 
ize? Do they have intra- or interspecific 
territories, and so forth. 

The taxonomic solution adopted here, 
pending further field work to clarify the 
situation, is that birds of the alpina species- 
group constitute a superspecies, with two 
component semispecies, nomenclaturally 
treated as species: M. alpina (including 
grisea), living in the wetter Andes of Co- 
lombia, Ecuador, Peru, and northwestern 
Bolivia (see Fig. 18), and M. cinerea (in- 
cluding argentina) living in the drier 
Andes of central and southern Bolivia, 
northwestern Argentina, and central Chile 
(see Fig. 18). 

5. The albifrons Species-Group 
The albifrons Superspecies 

Muscisaxicola albifrons. Muscisaxicola 
albifrons is usually found at higher alti- 
tudes than are most other species of the 
genus, except perhaps jiininensis. In Chile, 



224 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



Goodall et al. (1957) give the altitudinal 
range from 4000 to 5000 meters. Their 
photograph (page 163) shows the habitat 
of the species in Chile: gently undulating 
slopes with sparse and overgrazed scrub 
and bunch-grass. In Peru and Bolivia I 
saw the species on valley slopes where 
grazed grasslands were interrupted by 
cliffs and boulders, between 4300 and 4500 
meters. M. alhifrons ranges from northern 
Peru to extreme northern Chile and Bolivia; 
it has not been reported from northwestern 
Argentina. As far as is known it is a resi- 
dent species. There is no geographic vari- 
ation in this species. 

Miiscisaxicolo flaviniicha. Muscisaxicola 
flavimicha is a southern South American 
species ranging from central Chile and 
adjacent Argentina to Tierra del Fuego. It 
occurs up to about 4000 meters, according 
to Goodall et al. (1957), but in the Andes 
south of about 40° S its altitudinal limits 
are lower, because the Andes themselves 
rarely exceed 3000 to 3500 meters. The 
habitat of this species consists of barren 
rocky slopes with small cliffs or boulders, 
and very little herbaceous vegetation. 
Goodall et al. (1957) say that fhvimicha 
occurs especially near streams and moun- 
tain lakes, but in my experience, it also 
breeds in regions devoid of either streams 
or lakes. 

M. flavimicha is migratory and goes as 
far north as northern Peru (Libertad) to 
spend the southern hemisphere winter. 
This migration seems to take place entirely 
at high altitudes in the Andes. 

There is minor geographical variation in 
M. flavimicha. Olrog (1949b) described 
the birds of Tierra del Fuego as a distinct 
subspecies {brevirostris) because they have 
smaller bills and wings, and are darker in 
color than mainland birds. From an evo- 
lutionary point of view, M. flavimicha has 
one weakly differentiated isolate on Tierra 
del Fuego. It should be pointed out that 
although the type locality of brevirostris is 
Ushuaia (Tierra Grande, Tierra del 
Fuego), Olrog (1949b) did not think that 



these birds bred on this island, but rather 
on the mountains of Navarino and Hoste 
Islands and in the Cape Horn archipelago. 
These populations are probably rare or un- 
common. It was not until 1928 that 
Reynolds ( 1934 ) observed and collected 
this species in Harberton, Beagle Canal. 
Later (1935) he reported that he might 
have seen it on Freycinet. 

Muscisaxicola frontalis 

This species breeds in Chile (Anto- 
fagasta) and Argentina (Mendoza) in the 
north and from there southward to the 
latitude of Colchagua and Rio Negro 
(Johnson, 1967). Bond and de Schauensee 
(1942: 340) mentioned four specimens 
from Viloca and La Cumbre in Bolivia, and 
Bond (1947: 129) cited one male collected 
21 April at Pampa de los Arrieros near 
Arequipa in southern Peru. I agree with 
Bond that these birds are migrants from 
the south. The breeding habitat of M. 
frontalis seems to be stony and rocky 
slopes with sparse vegetation of small 
shrubs and bunch-grass. During the non- 
breeding season, the species occurs also at 
very high altitudes, either in bunch-grass 
steppes or in rocky spots near marshy areas 
(personal observation). M. frontalis does 
not show any detectable geographic vari- 
ation. 

DISCUSSION 

The bush and ground tyrants analyzed in 
Part II show clear-cut instances of various 
stages of the speciation process. Thus, at 
least the most recent events in the adaptive 
radiation of these birds can be traced. Yet, 
the earlier episodes of their evolutionary 
history appear shrouded in uncertainty. 
Indeed, these flycatchers have a rather 
large number of both taxonomically iso- 
lated species, and taxonomically closely 
related but extensively sympatric species. 

Table 4 shows this evolutionary dichot- 
omy. Of thirty-three species and semi- 
species, nine (about 27 percent) possess 



Evolution of Ground Tyrants • Smith and Vuilleumier 225 



Table 4. Stages of the specl^tion process in the Bush and Ground Tyrants 



Geographical variation 



Species and 
semispecies 



Absent 



Gradual 



Discontinuous 
( Isolates slightly- 
differentiated ) 



Member 

of a 
Super- 
species 



Second- 
ary 
contact 



No 

close 

relative 



Closest 

relative 

broadly 

sympatric 



Xolmis striaticollis — 

X. pernix — 

X. fumigata + 

X. juscorufa + 

X. signaiar + 

X. enjthropijgia — 

X. rufipeimis + 

X. pyrope — 

X. cinerea — 

X. velata — 

X. coronata — 

X. dominicana — 

X. irupero — 

X. rubetra + 

Neoxohnis rufiventris + 

Agriornis inoniana — 

A. albican da — 

A. livida — 

A. microptera — 

A. mtirina + 

Muscisaxicola brevicatida — 

M. macuUwstris — 

M. fluviatilis 

M. macloviana 

M. capistrata 

M. nifivertex 

M. junmensis 

M. albilora 

M. alpina 

M. cinerea 

M. albifrons 

M. fJavinucha 

M. frontalis 

Totals: 33 



+ 

+ 

+ 
+ 

+ 

+ 
13 



+ 



+ 



+ 
+ 



+' 



+ 



+ ? 
+ 



+ 
+ 



+ 

+? 


+ 




+* 











+ 






+ 


+ 


— 






+ 






+? 


_ 






+ 


+ 


i + 

i + 
( + 


+' 
+' 


— 


+ 


j + 




8 ^ 


9 


9 


4 



+ 

+ 



6 



+ 



+ 
+ 

+ 
+ 
+ 
+ 
+ 



+ 
+ 



+ 
+ 
+ 



+ 
14 



1 X. striaticollis shows what appears to be a stepped cline in southern Peru and northern Bolivia. The interpretation as 
"secondan.- contact" is hypothetical. 

^ X. signata may or may not belong in Xolmis. 

2 X. pyrope shows gradual geographical variation on the continent, but has a subspecifically distinct isolate on Chiloe 
Island. 

* A. montana shows a phenomenon similar to that of X. striaticollis above. 

^ The two semispecies of the M. alpina superspecies appear to overlap in a very- narrow zone of the Bolivian Andes. 

Semispecies belonging in the same superspecies are enclosed in braces. 



morphologically differentiated isolates, nine 
(about 27 percent) are members of super- 
species, and four (about 12 percent) have 
what appear to be secondary contacts sug- 
gesting former isolation. At the same time, 
of thirty-three species (regardless of 
whether they also show active speciation 



or not), SLX (about 18 percent) are taxo- 
nomicallv isolated, and fourteen (about 42 
percent) are broadly sympatric with their 
most closely related congener(s). These 
two aspects of the evolutionary' history of 
these flycatchers are analyzed further 
below. 



226 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



Presently Active Speciation 

One of the most important events of 
geographical speciation is the estabhshment 
of isolates within a species' range (see, e.g., 
Mayr, 1963). The corollaries of geograph- 
ical isolation are the cessation of gene flow 
between populations, and the subsequent 
initiation of genetic divergence in the 
isolates under different selection regimes. 
The presence of genetically distinct iso- 
lates is therefore an index to the speciation 
potential of extant species (Prakash, Le- 
wontin, and Hubby, 1969). 

In the absence of actual data on the 
genetics of populations of Andean birds, 
including the Tyrannidae of the present 
study, I have of necessity relied on morpho- 
logical differentiation to make statements 
about genetic differentiation. My con- 
clusions about speciation are therefore 
based on the assumption that morpho- 
logical differentiation between two isolated 
populations is positively correlated with 
their genetic differentiation. That this is 
indeed the case is suggested by the work 
of Hubby and Throckmorton (1968) on 
isozyme variability among triads of closely 
related species of Drosophila. 

Nevertheless, genetic differentiation can 
be extensive even in the absence of parallel 
morphological variation. Thus the bias of 
my study would be to underestimate 
genetic differentiation: first, by consider- 
ing that morphologically similar isolates 
are also genetically similar, and second, by 
failing to recognize minor, but real, geo- 
graphical variation. 

The first source of bias may prove very 
difficult to erase. The second, however, 
could be eliminated by undertaking ex- 
haustive studies of geographical variation 
of each of the species treated here, espe- 
cially, perhaps, those considered to have 
no, or only gradual, variation (see Table 
4). As every systematist knows, such 
studies are extremely time consuming. 
Since my goal was to obtain broad com- 
parative data on speciation from a large 



sample of the entire Andean avifauna (see 
Vuilleumier, 1969b), I have deliberately 
chosen breadth at the expense of depth, 
and can only hope that the many problems 
uncovered will be taken up as separate 
projects by others. 

The observed instances of presently 
active speciation fall in two categories: 
(a) differentiation centered around more 
or less well-marked geographical and eco- 
logical barriers, and (b) differentiation 
and secondary contacts not centered around 
such barriers. 

Geographical Barriers 

In two species and one superspecies of 
Xolmis, one species of Agriornis, and four 
species of Miiscisaxicoki, morphological 
differentiation can be correlated with the 
presence of barriers. In all these cases the 
isolates have been formally recognized 
taxonomically by previous ornithologists, 
usually as subspecies, but as species in the 
X. fumigata superspecies. 

From an evolutionary point of view, the 
barriers isolating these populations would 
therefore seem to inhibit gene flow from 
other populations, as evidenced by the 
moiphological differentiation. These bar- 
riers are listed below in geographical 
sequence from north to south (several of 
them have already been cited in Vuilleu- 
mier, 1968, 1969ai 1969b). 

( 1 ) Cesar Depression. The lowlands sepa- 
rating the Santa Marta Mountains of north- 
ern Colombia from the Perija Range seem 
to affect Xolmis pernix and X. fnmigato, 
two species or semispecies of the X. fumi- 
gata superspecies (see Fig. 12). In this 
case, morphological differentiation is suf- 
ficient to suggest that species level has 
been reached by the populations on either 
side of the gap. In Xolmis erytluopygia, 
a population appears restricted to the Santa 
Marta Range; it has been separated taxo- 
nomically as a subspecies {orinoma). The 
species is absent from the Perija Range, 
and also from the remainder of the Colom- 



Evolution of Ground Tyrants • Smith and Vuillenmier 227 



bian Andes, except in the extreme south 
(Nariho). Whether this patchy distri- 
bution means that extinction has occurred 
in the Andes between Nariiio and Santa 
Marta or whether it is an artifact reflecting 
insufficient collecting (as a result of the 
rarity of the species?) cannot be resolved 
at the present time. Xohnis striatic oil is has 
an apparently undifferentiated population 
in the Santa Marta Mountains. 

(2) Rio Torhes V alley. This arid valley 
isolates the Andes of eastern Tachira, 
Merida, and Trujillo from those of western 
Tachira in Venezuela. The populations of 
Xohnis fumigata living in the Andes to 
the east of this valley are clearly dis- 
tinguishable moi-phologically from other 
populations, and have been separated taxo- 
nomicallv as the subspecies liigubris (see 
Fig. 12)'. 

(3) Rio Magdalena Valley. The Central 
and Eastern Andes of Colombia are iso- 
lated from each other by this dry valley. 
Muscisaxicola alpina has an isolated popu- 
lation in the Central Andes ( taxonomically 
separated as the subspecies colomhianu) 
and another in the Eastern Andes (sub- 
species qttesadae) (see Fig. 18). The mor- 
phological differentiation between these 
populations is slight. 

(4) Upper Rio Magdalena Valley and 
Soidhwestern End of Eastern Andes. The 
southwestern part of the Eastern Andes 
appears too low for paramo vegetation to 
occur, so that there is a distribution gap 
between the Andes of Nariiio in southern 
Colombia and the central part of the 
Eastern Andes. Muscisaxicola maculirostris 
has a well-differentiated isolate ( subspecies 
niceforoi) in the Eastern Andes, isolated 
by a very large gap (marked by ? in Fig. 
15) from the Ecuadorean population (sub- 
species rtifescens) . The problem of whether 
or not the actual barrier corresponds to a 
gap in the vegetation preferred by this 
species in Colombia remains to be estab- 
lished in the field, hence my ? in Figure 
15. A similar gap exists in the distribution 



of Muscisaxicola alpina (marked with a ? 
in Fig. 18). 

(5) Upper Rio Maraiion Valley and North- 
ern Peruvian Loiv. In the Andes of north- 
em Peru, the Rio Maraiion Valley cuts 
deeply across the relatively low-lying 
mountains as its course changes from a 
south-north to a west-east direction. The 
Andes west of this bend are especially low, 
and are probably unsuitable for many high 
altitude birds, because the upper vegeta- 
tion zones, especially those above the 
timber line, are either absent or of very 
small area. This complex barrier separates 
populations of Xolmis fumigata (see Fig. 
12), with slight morphological differenti- 
ation only. In Muscisaxicola, however, two 
species have well-marked populations on 
either side of this gap. In M. maculirostris, 
the subspecies rufescens occurs north of 
the barrier, and maculirostris sovith of it 
(see Fig. 15). In M. alpina, the northern 
subspecies is alpina and the southern grisea 
(see Fig. 18). Finally, Agriornis montana 
has weakly differentiated populations sepa- 
rated by this gap (see Fig. 14), taxo- 
nomically recognized as subspecies solitaria 
(north) and insolens (south). Thus the 
Maraiion gap has various effects on fly- 
catchers living in this area, but is it note- 
woithy that none of the populations so 
isolated belong to a superspecies. 

(6) Rio Apurimac Valley. The upper 
Rio Apurimac and some of its tributaries 
(Rio Pampas, Rio Mantaro) may constitute 
a barrier between Xolmis fumigata and X. 
fuscorufa of the X. fumigata superspecies 
(see Fig. 12), and between the weakly 
marked subspecies erythropygia and boli- 
viana of Xolmis erythropygia. In both 
cases, however, much more collecting is 
necessary before the precise barrier area 
between taxa can be ascertained. 

( 7 ) Andes of Western Argentina. The huge 
hiatus (over 1000 km) in the distribution 
of Agriornis microptera (see Fig. 13) is 
difficult to interpret in terms of a single 



228 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



Table 5. Small peripheral isolates that show differentiation in morphological characters 



Species or suijerspecies 
having isolate 



Xolmis fumigata 
Xolmis fumigata 
Xolmis enjthiopijgia 
Xolmis ptjrope 
Agriornis moniana 
Muscisaxicola maculiwstris 
Muscisaxicola maciilirostris 
Muscisaxicola macloviana 
Muscisaxicola alpina 
Muscisaxicola alpina 
Muscisaxicola flavinucha 



Geographical isolate 
( taxonomic rank ) 



Location of isolate 



liigubris ( subspecies ) 
pernix ( species or semispecies ) 
oriiio))ia ( subspecies ) 
fortis ( subspecies ) 
solitaria ( subspecies ) 
niccforoi (subspecies) 
rufescens ( subspecies ) 
macloviana ( subspecies ) 
Columbiana ( subspecies ) 
quesadae (subspecies) 
brevirostris ( subspecies ) 



Merida, Venezuela 

Santa Marta, Colombia 

Santa Marta, Colombia 

Chiloe Island, Chile 

Ecuador and southern Colombia 

Eastern Andes, Colombia 

Ecuador 

Falkland Islands 

Central Andes, Colombia 

Eastern Andes, Colom1)ia 

Tierra del Fuego 



"barrier." The habitats favored by this 
species seem to be zones of xeric, rocky 
plains, which are largely missing in the 
Andes, and the foothills between the ranges 
of the subspecies ondecola (high Andes) 
and microptera (Patagonia). Range dis- 
junctions similar to the one found in A. 
microptera are also observed in Pterocne- 
mia pennata, Tinamotis, and Diiica. 

(8) Straits Between Chiloe Island and 
Mainland. The Chiloe Island population 
of Xolmis pyrope appears slightly differen- 
tiated from the mainland ones. 

(9) Straits of Magellan. The population of 
Muscisaxicola flavinucha living on Tierra 
del Fuego or on islands south of the Beagle 
Canal may be slightly differentiated 
morphologically. More field work is needed 
to check whether the variation is truly 
discontinuous (and if so, whether the 
barrier is indeed the Straits of Magellan) 
or merely clinal. 

(10) Straits Separating Falkland Islands 
From Mainland. The Falkland population 
of Muscisaxicola macloviana is slightly 
differentiated from the mainland ones (see 
Fig. 17). 

From an ecological standpoint, the ten 
gaps listed above are quite variable. Aside 
from barriers 8-10, which are marine, all 
others are terrestrial. Barriers 2, 3, 5, and 



6 may be roughly equivalent ecologically. 
They are situated around arid valleys cut- 
ting across high mountain ranges. Barrier 1 
is a more varied lowland gap. Finally, 
barriers 4 and 7 are ill-defined interrup- 
tions in the continuity of certain habitats. 
Of the moqohologically differentiated 
isolates separated by one or another of 
barriers 1-10, it is interesting to note that 
eleven may be considered peripheral and 
of small size (see Table 5). These isolates 
may have originated as small founder 
populations that colonized suitable habitats 
at the peripheiy of the range of the species 
(see Mayr, 1954). Eight of the eleven 
isolates occur north of barrier 5, in the 
Andes of Ecuador, Colombia, and Vene- 
zuela, where the distribution patterns of 
high altitude habitats, especially paramos, 
are quite insular (see Vuilleumier, 1970). 
The potential importance of peripheral 
isolates as incipient species has been em- 
phasized by Mayr (1963). Recently, Prak- 
ash, Lewontin, and Hubby (1969) have 
demonstrated that genetical differentiation 
is marked in the small and isolated periph- 
eral population of Drosophila pseudooh- 
scura in Colombia. 

Absence of Barriers 

I have argued that in Xolmis striaticoUis 
and Agriornis montana (see Fig. 14), and 



Evolution of Ground Tyrants • Smith and Vuilleumier 229 



in the Muscisaxicola alpina superspecies 
(see Fig. 18), secondary contacts are re- 
sponsible for the observed patterns of dis- 
tribution and variation: hybridization or 
secondary intergradation, and narrow mar- 
ginal overlap. Although these interpreta- 
tions need to be substantiated by more 
evidence, it is possible to state here that 
these contacts do not seem to take place 
in regions that can be called barriers. All 
three contact zones are located in the 
Andes between southern Peru and north- 
western Bolivia. 

In both the Muscisaxicola alhilora and 
M. alhifrons superspecies, a distribution 
gap appears to exist between the pairs of 
species or semispecies in the Andes of ex- 
treme northern Chile and of western 
Bolivia. Whether these gaps are real or 
artificial, resulting from the lack of inten- 
sive collecting in these areas is, of course, 
not possible to ascertain at the present 
time. In the alhilora superspecies, the two 
semispecies (or species) are quite similar 
morphologically, and secondary contact 
might conceivably result in hybridization. 
In the alhifrons superspecies, however, the 
differences between the two taxa are well 
marked, and contact might result in sym- 
patry (as it apparently does in the alpina 
superspecies). 

All of the supposed instances of second- 
ary contacts rest, unfortunately, on too 
few specimens having sufficient label data 
and on too few field data. The best field 
evidence, in the M. alpina superspecies, 
comes from my own exploratory field work, 
so I am better aware than anyone else of 
the limitations of the speculations ad- 
vanced here. 

Older Speciation Patterns 

The taxa of bush and ground tyrants 
that do not show evidence of active 
speciation fall in two categories: (a) 
taxonomically isolated species, and (b) 
taxonomically related, sympatric species. In 
the first group belong Xolmis pyrope, X. 



ruhetra, Agriornis murina, Neoxolmis rufi- 
ventris, and Muscisaxicola hrevicauda. 
Some of these species (e.g., Xolmis pyrope, 
Neoxolmis rufiventris, and Muscisaxicola 
hrevicauda) occupy relatively specialized 
habitats, as I pointed out in Part I. Nothing 
can be inferred of the history of these 
species on present evidence. 

In the second category (sympatric con- 
geners ) belong especially the species of the 
Xolmis fumigata, erythropygia, and cinerea 
species-groups, those of the Agriornis mon- 
tana and livida species-groups, and the 
species of the subgenus Muscisaxicola. 

Sympatry in Xolmis 

If the Xolmis fumigata superspecies is 
considered as one geographical species, it 
is broadly sympatric with X. striaticollis, 
X. erythropygia, and X. rufipennis. These 
species show various degrees of habitat 
differentiation, so that habitat co-occu- 
pation may only occur between X. fumigata 
and X. erythropygia, whereas X. striaticollis 
and X. rufipennis are ecologically more 
distinct. The precise degree of habitat 
overlap between these species, however, is 
a problem needing future field work. In 
any event, this extensive sympatry makes 
it impossible to deduce anything about 
past speciation within the Andean species- 
groups of Xolmis. 

In the cinerea species-group, sympatry 
is again too extensive for inferences to be 
drawn about the history of the group in 
extra-Andean open habitats. 

Sympatry in Agriornis 

Two species of Agriornis (montana and 
alhicauda) are not only sympatric over 
large areas of the high Andes from Ecua- 
dor to northwestern Argentina, but they 
can also be found in similar habitats, at 
least in Bolivia (personal obser\^ation ) . 
Furthermore, A. microptera is sympatric 
with both montana and alhicauda in parts 
of the altiplano of Bolivia and northwestern 
Argentina. 



230 Bulletin Mmcinn of Comparative Zoology, Vol. 141, No. 5 



Sympatry in Muscisaxicola 

The only species of the subgenus Musci- 
saxicoJa that lives in complete allopatr\' is 
M. fluviatilis, a lowland species. In the 
uplands and high Andes, especially south 
of central Peru, up to four or five species 
of this subgenus live sympatrically. 

Sympatry in Mtiscisaxicola has long been 
known to ornithologists, yet no more than 
passing mention of the phenomenon can 
be found in the literature. As a result, the 
following discussion is compiled from my 
own field notes. 

Altitudinal and ecological replacement 
is obvious in Ecuador between M. maciiU- 
rostris, a species that lives in xeric steppes 
and brushlands in the inter- Andean region 
from about 2000 to 3700 meters, and M. 
alpina, a larger species that occurs in 
moister habitats such as meadows and 
bunch-grass from about 3500 to well above 
4100 meters. 

In central Peru, M. maailirostris ranges 
from about 1500 to 4000 meters, and 
broadly overlaps with M. riifivertex, which 
occurs from about 3000 to 4200 meters. At 
altitudes above 4000 meters in the altiplano 
of Peru and Bolivia, M. alpina (or M. 
cinerea), M. juninensis, M. albifrons, and 
locally also M. maculirostris and M. riifi- 
vertex occur together. Of the five species 
and superspecies, Af. rufivertex seems to 
occur lower down, whereas M. alJnfrons is 
found only at the highest altitudes. I have 
seen and collected M. alpina and M. junin- 
ensis (or M. cinerea and M. juninensis) in 
the same habitats in Bolivia, and have also 
observed these two species with M. albi- 
frons. In some localities, I have seen M. 
cinerea and M. rufivertex together, in 
others, M. cinerea and M. maculirostris. 
Thus true habitat co-occupancy involves 
usually two, but sometimes three, of the 
five species listed above. 

Further study of these sympatric associ- 
ations should be made to determine 
whether these species differ in habitat 
requirements, foraging sites, nesting habits, 



or in other details that might reduce 
competition among them. It is noticeable 
that the two semispecies of the alpina 
superspecies (alpina and cinerea) overlap 
in body size and bill size with juninensis 
and rufivertex. Thus food size partitioning 
may not be an important way of effecting 
ecological segregation. 

This preliminary discussion of sympatry 
in Xolmis, Agriornis, and Muscisaxicola 
shows that this phenomenon is quite ex- 
tensive in these genera. It would therefore 
seem that they have been present in the 
high Andes and the extra-Andean open 
regions of central South America for a long 
time, so long, in fact, that the early stages 
of their adaptive radiation have been 
obliterated by complex patterns of second- 
ary sympatry. 

LITERATURE CITED 

Bennett, A. G. 1926. A list of the birds of the 

Falkland Islands and dependencies. Ibis, ser. 

12, 2: 306-333. 
Berlepsch, H. von. 1907. Studien iiber Tyran- 

niden. Ornis, 14: 463-493. 
Bond, J. 1947. Notes on Peruvian Tyrannidae. 

Proc. Acad. Nat. Sci. Philadelphia, 99: 

127-154. 
Bond, J., and R. M. de Schauensee. 1942. The 

birds of Bolivia. Part I. Proc. Acad. Nat. 

Sci. Philadelphia, 94: 307-391. 
Carriker, M. a., Jr. 1932. Descriptions of new 

birds from Peru and Bolivia. Proc. Acad. 

Nat. Sci. Philadelphia, 87: 343-359. 
Cawkell, E. M., and J. E. Hamilton. 1961. 

The birds of the Falkland Islands. Ibis, 

103a: 1-27. 
Chapman, F. M. 1927. Descriptions of new 

birds from northwestern Peru and western 

Colombia. Amer. Mus. Novitates, No. 250: 

1-7. 
Cott, H. B. 1947. The edibility of birds. Proc. 

Zool. Soc. London, 116: 371-524. 
Cravvshay, R. 1907. The Birds of Tierra del 

Fuego. London: Bernard Quaritch, 158 pp. 
CuELLo, J., and E. Gerzenstein. 1962. Las 

aves del Uruguay. Lista sistematica, distri- 

bucion y notas. Com. Zool. Mus. Monte- 
video, 6(93): 1-191. 
DoRST, J. 1955. Recherches ecologiques sur les 

oiseaux des hauts plateaux peruviens. Trav. 

Inst. Fr. £tud. Andines, 5: 83-140. 



Evolution of Ground Tyrants • Smith and Viiilleumier 231 



. 1962. A propos de la nidification hypogee 

de quelqiies oiseaiix des hautes Andes 
peruviennes. Oiseau Rev. Fr. Ornithol., 32: 
5-14. 

Franze, B. 1927. Die Niederschlagsverhaltaisse 
in Siidamerika. Petermanns Geog. Mitt., 
Gotha, Ergiinzungsheft, No. 193, 79 pp. 

Friedmanx, H. 1927. Notes on some Argentina 
birds. Bull. Mus. Comp. Zool., 68(4): 139- 
236. 

GooDALL, J. D., A. W. Johnson, and R. A. 
Philippi B. 1957. Las aves de Chile, su 
conociniiento y sus costnmbres. I. Buenos 
Aires: Piatt Establecimientos Graficos S. A., 
441 pp. 

Hellmayr, C. E. 1927. Catalogue of birds of 
the Americas and the adjacent islands in 
Field Museum of Natural History. Part V. 
Tyrannidae. Field Mus. Nat. Hist., Publ. 
242, Zool. Sen, 13: 1-517. 

. 1932. The birds of Chile. Field Mus. 

Nat. Hist., Publ. 308, Zool. Ser., 19: 1-472. 

Hubby, J. L., and L. H. Throckmorton. 1968. 
Protein differences in Drosophila. IV. A 
study of sibling species. Amer. Naturalist, 
102: 193-205. 

Hudson, W. H. 1920. Birds of La Plata. Vol. 1. 
New York: E. P. Dutton, 244 pp. 

Johnson, A. W. 1967. The Birds of Chile and 
Adjacent Regions of Argentina, Bolivia and 
Peru. Vol. II. Buenos Aires: Piatt Estableci- 
mientos Graficos S. A., 447 pp. 

Knock, K. 1930. Klimakunde von Siidamerika, 
Handbuch der Klimatologie, II, Teil G., Ber- 
lin, Gebriider Borntraeger. 

KoEPCKE, M. 1958. Die Vogel des Waldes von 
Zarate. Bonn. Zool. Beitr., 9: 130-193. 

. 1964. Las aves del Departamento de 

Lima. Lima: Morsom S. A., 128 pp. 

Lauer, W. 1952. Humide und aride Jahreszeiten 
in Siidamerika und Afrika und ihre Bezieh- 
ungen zu den Vegetationsgiirteln. ..Bonn. 
Geogr. Alihandl., 9: 15-98. 

Leveque, R. 1964. Notes on Ecuadorian birds. 
Ibis, 106: 52-62. 

Maclean, G. L. 1969. The nest and eggs of the 
Chocolate Tyrant Neoxolmis rufiventris 
(Vieillot). Auk, 86: 144-145. 

Marchant, S. 1960. The lireeding of some S. W. 

Ecuadorian birds. Ibis, 102: 349-382. 
Mayr, E. 1942. Systematics and the Origin of 
Species. New York: Columbia University 
Press, 334 pp. 

. 1954. Change of genetic environment 

and evolution. In J. Huxley, A. C. Hardy, 
and E. B. Ford, eds.. Evolution as a Process. 
London: Allen and Unwin, pp. 157-180. 
. 1959. Trends in a\ian systematics. Ibis, 



. 1963. Animal Species and Evolution. 

Cambridge, Massachusetts: Belknap Press, 
Harvard University Press, 797 pp. 

Morrison, A. 1939. The birds of tlie Depart- 
ment of Huancavelica, Peru. Ibis, ser. 14, 3: 
453-486. 

NiETHAXCMER, G. 1956. Zur Vogelwelt Bohviens. 
Teil II. Passeres. Bonn. Zool. Beitr., 7: 
84-150. 

Olrog, C. C. 1948. Observaciones sobre la avi- 
fauna de Tierra del Fuego y Chile. Acta 
Zool. Lilloana, 5: 437-531. 

. 1949a. La avifauna del Aconquija. Acta 

Zool. Lilloana, 7: 139-159. 

. 1949b. Notas ornitologicas sobre la 

coleccion del Institute Miguel Lillo. I. Acta 
Zool. Lilloana, 8: 209-217. 

. 1963. Lista v distribucion de las aves 



101: 293-302. 



argentinas. Opera Lilloana, 9: 1-377. 

Peters, J. L., and J. A. Griswold. 1943. Birds 
of tlie Hai-vard Peru\'ian expedition. Bull. 
Mus. Comp. Zool., 92: 281-327. 

Philippi B., R. A. 1964. Catalogo de las aves 
chilenas con su distribucion geografica. In- 
vest. Zool. Chilenas, 11: 1-179. 

Pinto, O. M. de O. 1944. Catalogo das aves do 
Brasil. 2a. Parte. Departamento de Zoologia, 
Secretaria da Agricultura, Sao Paulo, 700 pp. 

Prakash, S., R. C. Lewontin, and J. L. Hubby. 
1969. A molecular approach to the study of 
genie heterozygosity in natural populations. 
IV. Patterns of genie variation in central, 
marginal and isolated populations of Drosoph- 
ila pscudoohscura. Genetics, 61: 841-858. 

Reynolds, P. W. 1934. Apuntes sobre aves de 
Tierra del Fuego. Hornero, 5: 339-353. 

. 1935. Notes on tlie birds of Cape Horn. 

Ibis, ser. 13, 5: 65-101. 

Schauensee, R. M. de. 1966. The species of 
birds of South America and their distribution. 
Narberth, Pennsylvania: Li\ingston Publish- 
ing Company, 577 pp. 

Simpson, G. G. 1941. Large Pleistocene felines 
of North America. Amer. Mus. Novitates, 
No. 1136: 1-27. 

. 1953. The Major Features of Evolution. 

New York and London: Columbia University 
Press, 434 pp. 

Taczanowski, L. 1874. Liste des oiseaux re- 
cueillis par M. Constantin Jelski dans la partie 
centrale du Perou occidental. Proc. Zool. Soc. 
London, 1874: 501-565. 

Todd, W. E. C, and M. A. Carriker, Jr. 1922. 
The birds of the Santa Marta region of 
Colombia: a study in altitudinal distribution. 
Arm. Carnegie Mus., 14: 1-611. 

VuiLLEUMiER, F. 1965. Relationships and evo- 



232 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



lution within the Cracidae. Bull. Mus. Comp. 

Zool., 134: 1-27. 
— . 1968. Population structure of the Asthenes 

flammulata superspecies (Aves: Furnariidae ) . 

Breviora, No. 297: 1-21. 
. 1969a. Systematics and evolution in 

Diglossa (Aves: Coerebidae). Amer. Mus. 

Novitates, No. 2381: 1-44. 
. 1969b. Pleistocene speciation in birds 

living in tlie high Andes. Nature (London), 

223: 1179-1180. 
. 1969c. Field notes on some birds from tlie 

Bolivian Andes. Ibis, 111: 599-608. 
. 1970. Insular biogeography in continental 

regions. I. The northern Andes of South 

America. Amer. Naturalist (in press). 



Wetmore, a. 1926a. Observations on the birds 

of Argentina, Paraguay, Uruguay, and Chile. 

Bull. U. S. Nat. Mus., 133: 1-448. 
. 1926b. Report on a collection of birds 

made by J. R. Pemberton in Patagonia. Univ. 

Calif. Pulil. Zool., 24: 395-474. 
. 1946. New birds from Coloml^ia. Smith- 



son. Misc. Coll., 106: 1-14. 

Wilson, E. O., and W. L. Brown. 1953. The 
subspecies concept and its taxonomic ap- 
plication. Systematic Zool., 2: 97-111. 

ZiMMEB, J. T. 1937. Studies of Peruvian birds. 
No. 26. Notes on the genera Agriornis, 
Muscisaxicola, Myiotheretes, Ochthoeca, Co- 
lonia, Knipolegus, Phaeotriccus, Fluvicola, 
and Ramphotrigon. Amer. Mus. Novitates, 
No. 93(): 1-27. 



Chapter II. Behavior of Muscisaxicola and Related Genera 

W. JOHN SMITH 

Abstract. Terrestrial species of tyrannic! flycatchers in tlie genera Muscisaxicola, Agriornis, and 
Xolmis appear to be closely related to the less terrestrial species of Xolmis and to the essentially nonter- 
restrial genera Ochthoeca, Pywcephalus, and Sayornis on the basis of characteristics assessable in mu- 
seum skins. Although the ecological span of the group as a whole is great, specific habitat preferences, 
foraging methods, patterns of locomotion and nesting behavior all vary along continua that transcend 
generic limits. Further, members of all of the genera are shown to employ in similar ways displays that 
are alike in various aspects of form, and tliat do not appear to have converged as the result of selection 
pressures operating in the semi-open to open habitats. It is concluded that all of the species, both ter- 
restrial and nonterrestrial, belong to one phylogenetic group. 



TABLE OF CONTENTS 

Introduction 233 

Field Sites 235 

Acknowledgments 236 

Part I: Nondisplay Behavior 236 

Habitat Preferences 236 

( 1 ) Muscisaxicola 236 

(2) Xohim and Neoxolmis 237 

(3) Agriornis 238 

(4) Ochthoeca 239 

Locomotion and Foraging Behavior ^, 239 

( 1 ) Flight and Aerial Foraging 239 

(2) Perching and Locomotion and For- 
aging on the Ground 239 

Flocking 24 1 

Nest Construction 242 

Part II: Display Behavior 242 

11a. Muscisaxicola 242 

lib. Xolmis 255 

lie. Agriornis 258 

lid. Ochthoeca ..___ 259 

Comparative Sununary and Discussion — 261 

Conclusions __ 267 

Literature Cited 268 

Bull. Mus. Comp. 



INTRODUCTION 

The markedly terrestrial genus Musci- 
saxicola is one of the most extreme products 
of the South American radiation of the 
Tyrannidae. It is not the only genus com- 
mitted to ground-dwelhng, but its trim, 
longlegged, Oenonthe-like species are a 
close-knit group, strikingly different in 
aspect from better known tyrannids such 
as the kingbirds (Tyrannus), which perch 
on vegetation and forage largely by pursuit 
of aerial prey. A superficial examination 
of museum skins supports the usual ar- 
rangement of taxonomic lists which places 
Xolmis (herein including Myiotheretes) 
and Agriornis close to Muscisaxicola and 
suggests relationship to nonterrestrial 
genera such as Ochthoeca, Sayornis, and 
Pyrocephalus. The purpose of this paper 
is first to describe behavior — primarily 
foraging behavior, habitat preferences, and 
display (stereotyped communicative be- 
havior) — in all but the last two genera, 
which are reported upon elsewhere ( Smith, 
1969, in press, 1 and 2, and 1967). 
Secondly, it is to show that this information 

Zool., 141(5): 233-268, March, 1971 233 



234 Bulletin Museum of Comparotice Zoology, Vol. 141, No. 5 



supports the apparent phylogenetic re- 
lationship of all of these genera. 

As subjects for a comparative study of 
behavior patterns, the terrestrial tyrannid 
genera appear promising, largely because 
of their open habitats. Easily seen birds 
are more easily studied than are birds ob- 
scured by dense vegetation. Thus, when I 
began in 1959 a series of comparative be- 
havioral studies of a wide variety of tyran- 
nids, I included a program on this group. 
Of the many tyrannids with which I even- 
tually worked, however, these proved 
among the most difficult for reasons which 
had not been anticipated. 

In the case of Muscisaxicola, for instance, 
most of my field work was in the Andes 
of central Chile. In early spring, several 
species were grouped into loose flocks at 
accessible altitudes, frequenting widely 
scattered good foraging sites where melt- 
water collected and where a carpet of 
green vegetation was present. When higher 
slopes became more open, they moved up 
in pairs and scattered. In these sparse 
populations there was relatively little dis- 
play behavior. Most work in my first 
spring was on steep slopes where a bird 
could put itself out of sight by flying 50 
meters, leaving me scrambling to locate it 
again. For my second spring in Chile I 
found better sites, but had less time avail- 
able and lost most of that to an unseason- 
ably late snowstorm that closed access to 
higher altitudes and brought the birds 
back down into their loose flocks. 

An additional problem of work in early 
spring in central Andean valleys was high 
wind in at least the early morning hours 
when birds are most active. This wind 
made tape recording difficult, yet the re- 
cording was essential because accurate 
distinctions among the various calls in the 
repertoires of Muscisaxicola species cannot 
be made by the human ear. 

Similar problems were met in briefer 
attempts to study members of this genus 
in areas other than central Chile, although 



reasonably dense populations of Muscisaxi- 
cola macuUrostris were found in Ecuador, 
and of M. brevicauda in Peru. Within the 
other genera, both Xolmis pijropc and 
Agriornis livicla were much rarer as breed- 
ing birds in areas accessible from Santiago, 
Chile, than I had surmised from the litera- 
ture. Xolmis pijrope was more common on 
Tierra del Fuego, along with Muscisaxi- 
cola capistrata, although neither formed 
dense populations and both were in late 
phases of the breeding cycle when I was 
there. Again, Tierra del Fuego is almost 
continuously swept by high winds, hinder- 
ing tape recording. Elsewhere, in northern 
Argentina both Xolmis and Agriornis, and 
in southern Brazil Xolmis species were not 
found in dense populations suitable for 
study of their displays. It seems possible 
that at least Agriornis species may custo- 
marily have veiy thinly scattered popu- 
lations (.see Goodall et al., 1957). Finally, 
I spent little time in regions where 
Oclithoeca species occur. 

Under suitable conditions, there are ad- 
vantages to comparative behavioral studies 
of several species in one season. Yet, with 
relatively undemonstrative species at low 
population densities, this tactic yields in- 
complete information. The alternative of 
studying one or two species at a time re- 
quires considerable financial resources and 
many seasons. In the present case, where 
the end was taxonomic and the field work 
was, for the most part, expected to confirm 
conclusions derived from less expensive 
methods, large expenditure's did not appear 
justified. Accordingly, I de\'oted what 
field time I could allot to continuing a 
broad approach, even after initial experi- 
ence showed this would yield quite incom- 
plete results. Granted our nearly complete 
ignorance of these birds, even incomplete 
comparative information could be expected 
to provide useful perspective. I believe 
the results aid in making phylogenetic 
interpretations and justify the choice of 
tactic. It should be noted, however, that 



Evolution of Ground Tyrants • Smith and Vuilleumier 235 



the observed sample of display behavior is 
too small to be of more than limited use 
in studies of communication per se. 

The behavioral information is insuffici- 
ent, by itself, to support the taxonomic 
views implied by my assignments of spe- 
cies to genera herein. This information is 
not intended to stand alone, but to be 
considered with Vuilleumier's work (Chap- 
ter 1, which includes a formal taxonomic 
conclusion ) . 

This presentation is divided into two 
main parts; the first comprises nondisplay 
behavior: habitat preferences, foraging 
methods, methods of locomotion, and nest 
structure and placement. This is primarily 
to facilitate generic comparisons, and much 
detail about species has been omitted or 
stated briefly in a general form. While the 
data at hand would permit much more 
complete descriptions in some cases, useful 
species comparisons require even more in- 
formation, gathered under a greater variety 
of circumstances. 

The second part deals with display be- 
havior, and all available data judged to 
be useful are presented. Because these 
data are limited, however, formal names 
have been assigned to the vocal displays 
only provisionally, and by comparison with 
the more thoroughly studied genus Sayor- 
nis (Smith, 1969, and in press, 1 and 2). 
These names are often reduced after their 
first appearance to initials (e.g.. Simple 
Vocalization becomes SV)- Illustrations of 
vocalizations are based on analyses (Sona- 
grams ) made with a Kay Electric Company 
6061-A Sona-Graph. In all figures the 
ordinates represent frequency marked in 
kilocycles per second, and the abscissae 
duration in tenths of seconds. In order to 
illustrate better the detail in the relativelv 
brief vocalizations of the Muscisaxicola 
species, the temporal scales have been 
doubled (with corresponding halving of 
the frequency scales) with respect to the 
species of the other genera. 



FIELD SITES 

(1) Chile: 

Most sites were in the central Andes, in 
two regions: {a) Lagunillas, about 60 
kilometers east of Santiago, a relatively 
well-\'egetated region centering around a 
ridge at about 2300 meters; and {h) farther 
east at the headwaters of the Rio Yeso, a 
tributaiy of the Rio Maipo near the 
Argentine border; a relatively arid, rocky 
region. The two main sites on the Rio 
Yeso were the headwater lake at approxi- 
mately 2500 meters, and about 3 kilome- 
ters downstream on a gravelly shelf 
where a small tributary stream descends 
from Cerro Morado to the east. In ad- 
dition, one visit was made to sites at a 
similar altitude north of Santiago, Farel- 
lones and La Parva. 

Aoriornis livido was studied near Los 
Molles, Aconcagua Province, along the 
coast nordiwest of Santiago. Another 
coastal site was west of Santiago at Isla 
Negra, Valparaiso Province. A few ob- 
servations were made on the outskirts of 
Santiago at Fundo lo Recabarren on Cerro 
Manquehue. Other Chilean sites \\'ere on 
Ticrra del Fuego at the farm school Las 
Mercedes and at Cameron sheep station, 
respectively north and south of Bahia 
Inutil. 

(2) Argentina: 

Limited work was done in southern 
Tucuman Province, near La Cocha, and 
in the cordillera of western Tucuman above 
Tafi del Valle (about 2200 to 2500 meters) 
and in the pass of Inficrnillo ( 3040 meters ) . 
A brief stop was made for Xolmis irupero 
on a gravelly river plain (altitude about 
1400 meters) about 8 kilometers north of 
San Salvador de Jujuy. 

(3) Brazil: 

Observations were made in the interior 
plains of the state of Sao Paulo near Piras- 
sununga and Bebedoura, and in savannas 



236 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



of the Territory of Amapa near Porto 
Platon. 

(4) Peru: 

My main site was in cotton fields near 
the coastal town of Lurin, about 35 kilome- 
ters south of Lima. Less time was spent 
in a steep-sided gully ("quebrada") of a 
tributary of the Rio Rimac just west of 
Surco (about SO kilometers east of Lima 
in the arid foothills at less than 2000 
meters ) . 

(5) Ecuador: 

All sites were reached from Quito. The 
lowest were all within 5 kilometers of the 
town of San Antonio, an arid region on the 
equator about 25 kilometers north of Quito 
on the northern slopes of Cerro Pichincha. 
Another set of sites ranged from the pass 
between Cerro Pichincha and Cerro 
Ataeaso about 5 kilometers southward in 
the paramo zone along the east face of 
Ataeaso (over 3000 meters), as well as 
eastward, down from the pass about 4 
kilometers to the upper reaches of good 
montane forest. Finally, two mornings were 
spent south of Machachi in the paramo on 
the western slopes of \^olcan Cotopaxi, at 
3800 meters. 

ACKNOWLEDGMENTS 

I am deeply grateful to Dr. Patricio 
Sanchez, my host and closest colleague in 
Chile, for aiding enormously in all of the 
Chilean phases of my work and for teach- 
ing me much about Latin America. Similar 
thanks are due my hosts in other countries: 
Dr. and Mrs. Abraham Willink, Drs. H. W. 
and Maria Koepcke, Dr. Paulo Vanzolini, 
Father J. Sartori, Mr. and Mrs. M. Piggot. 
and Mr. and Mrs. J. Neusenger. In addi- 
tion, among many persons who were very 
helpful, the following deserve special 
thanks: Dr. Martin H. Moynihan, Dr. Ernst 
Mayr, Dr. J. Valencia, Dr. C. Doggen- 
weiler, Mr. J. D. Goodall, the late Dr. 
R. A. Philippi B.. Sr. Luis E. Pena, Mr. A. 
Hundey, Sr. T. Radonich, Dr. C. C. Olrog, 



Dr. A. Teran, Dr. D. Lancaster, Sr. P. 
Leahy, Sr. P. Autino, Dr. H. Camargo, Dr. 
and Mrs. A. S. Rand, and Mr. L. Freire. 
Finally, I should hke to thank my wife, 
Susan, who assisted with all aspects of the 
field work on mv final visit to Ecuador and 
with the preparation of the figures and 
manuscript. 

I am indebted to various sources for fi- 
nancial assistance. The work began with a 
brief visit to Ecuador financed in part 
from a grant from the Chapman Memorial 
Fund of the American Museum of Natural 
History. Subseciuently the project was 
supported bv National Science Foundation 
grants NSF-G 19261 and NSF-GB 2904. 
The following institutions have also pro- 
\'idcxl help in various ways: L^niversidad 
Catolica de Chile; Instituto Miguel Lillo 
de Tucuman, Argentina; Casa Humboldt 
in Lima, Peru; Secretaria da Agricultura in 
Sao Paulo, Brazil; Industria e Commercio 
de Mineros in Amapa, Brazil; Harvard 
University; and the University of Penn- 
sylvania. 

PART I. NONDISPLAY BEHAVIOR 

Habitat Preferences 

(1) MuscisaxicoJa. Among the most ter- 
restrial of all tyrannids, these birds inhabit 
open country in the Andean chain and 
southern tip of the continent (for a zoo- 
geographic account see Vuilleumier, Chap- 
ter I). Some (e.g., M. capistrafa, M. mac- 
loviano, and M. maculirostrls) usually fre- 
quent grass or other low vegetation, often 
among scattered bushes, while others (e.g., 
M. cinerea, M. fhivinucha, and M. aJhUora) 
breed in less well- vegetated areas, usually 
among rocks. Rocks or bushes in all eases 
offer the birds commanding perches up to 
a meter or more tall; sometimes riparian 
trees are used. 

The three smallest (M. maculirostris, M. 
hievicauda, and an eastern Andean one I 
have not seen, M. fhwiatilis) inhabit rela- 
tively low altitudes for the Andean species. 



Evolution of Ground Tyrants • Srnith and Vuilleumier 237 



Their habitats regularly include bushes and 
even trees. In coastal Peru, M. Ijrevicaiida 
lives in cotton fields, preferring sites with 
poor growth and foraging from the bushes 
or on the ground. Most of its natural 
habitats in Peru are also brushy ( Koepcke, 
1954, 1958), although on Ecuador's Santa 
Elena peninsula, Marchant (1960) found 
it in open countiy with or without bushes, 
as well as in "surprisingly thickly wooded 
places." 

M. viacidirostris inhabits the lowest alti- 
tudes of any Muscisaxicola species in the 
Chilean and Argentine Andes, along the 
upper edge of the zone of brushy vegeta- 
tion among bushes up to 2 meters tall 
(where these are not contiguous), and ex- 
tending above into rocky areas with fewer, 
much shorter bushes. In Tucuman Prov- 
ince, Argentina, the one pair found was 
in a grazed quebrada with few bushes and 
a riparian line of trees 8 to 10 meters tall; 
the birds perched within and on top of the 
trees. Ecuadorean sites varied from open, 
sparsely vegetated, steep, eroded slopes to 
brushy quebradas and fields of short com, 
comparable to the cotton fields of M. 
hrevicauda. 

In early spring, all Andean species seen 
in central Chile frequented upland mead- 
ows and river flats with short green 
vegetation and copious meltwater, or 
foraged in mud beside melting snow. Such 
sites are ephemeral, and through most of 
the breeding season, most species have 
xeric sites with dusty soil and sparse 
vegetation. I have seen M. cinerea, M. al- 
J)dora, M. macidirostris, and M. capistrata 
foraging on the banks and gravel bars of 
rivers and streams, and feeding by water 
may be common in other species (e.g., M. 
fluviatdis) . Goodall et al. (1957) remark 
that most Andean species occur near lakes, 
although they also occur elsewhere. 

Both Vuilleumier and I noted a tendency 
for dorsal coloration and/or saturation of 
color to match general characteristics of 
the surrounding habitat. Species inhabit- 
ing high rock deserts are grayest, those of 



lower, grassier areas are brown (grass is 
dry and brown through much of the time 
the birds are present). The two most 
southerly species and M. alpina in the wet 
paramo of Colombia and Ecuador are 
deepest and richest in color, and live in 
regions of relatively high rainfall. Similar 
color correlations are known from other 
tyrannids (e. g., Ttjronnus, Smith, 1966) 
and probably reduce the conspicuousness 
of birds which spend most of their time 
perched very much in the open. 

(2) Xolmis and Neoxolmis. (a) While ter- 
restrial by general tyrannid standards, most 
Xolmis species are much less so than are 
Muscisaxicola flycatchers. Most perch 
above the ground and live among bushes 
or trees commanding a view of grass or 
swamp vegetation. X. pyrope and the 
more northern species east of the Andes 
often inhabit edges of woods or such 
brushy regions as the Brazilian cerrados, 
although some, like X. velata, are found 
on more open savannas. Species in the 
Andean foothills will live in brushy stream 
cuts where the surrounding countryside is 
open. Even X. coronata, which winters on 
the pampas, "seldom" runs on the ground 
(Hudson, 1920: 142) and often perches on 
the tall grasses; this has been confirmed 
for individuals wintering near Rosario, 
Argentina, by Dr. Juan Delius (personal 
communication). Wetmore (1926) reports 
X. cinerea, X. coronata, X. irupero, X. 
pyrope, and X. ruhetra as occurring among 
and perching on bushes. 

Even the most terrestrial species, X. 
ruhetra and N. rufiventris, do not usually 
breed in fully open plains. Olrog (1959) 
gives the breeding habitat of the latter as 
shrubby open country ("campos arbusti- 
vos"). P. S. Humphrey (personal com- 
munication) found its principal breeding 
habitat to be "flat or gently rolling country, 
dotted with . . . shrubs about a foot or so" 
tall, and it was also common in rolling 
country with short grass tussocks. Maclean 
(1969) found the breeding habitat to be 



238 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



"flat, open, wind-swept pampa dotted with 
occasional Berheris bushes." 

There is less tendency for dorsal color- 
ation to match the habitat in this genus 
than in Muscisoxicolo. X. inipero, for in- 
stance, is a conspicuous, exposed, white 
bird. Others are more cryptic. For ex- 
ample, X. pijrope and X. cinerea are dark- 
gray and live in fairly wooded sites. X. 
velota is paler gray, overlaps with the 
darker X. cinerea, and tends to be in more 
open habitats. The most richly colored 
species, N. riifiventris, is again the south- 
ernmost. 

(b) I have had very little field experi- 
ence with members of the fiimigata species- 
group of Xolmis. Pairs of X. jumigata 
were seen in Ecuador where plantations 
of pine or eucaly]^5tus bordered pasture land 
on steep slopes at about 3000 meters. They 
perched in the lower limbs of trees or on 
telephone lines, often about 7 meters above 
the ground. Koepcke (1954) reports that 
X. striaticoUis prefers more densely vege- 
tated areas (around woods and small 
groups of trees) than do species of Agrior- 
nis and Miiscisaxicola in the western Andes 
above Lima. Todd and Carriker (1922) 
in the Santa Marta region of Colombia 
found this species from about 2500 to 3600 
meters, also along the edges of forest and 
in some more open sites with "stunted 
trees and shrubbery." Olrog (1959) says 
that X. striaticolUs inhabits brushy quebra- 
das and larger river valleys up to about 
2500 meters in northwestern Argentina. In 
Ecuador and Argentina, Vuilleumier found 
the species in "open sites, but in Ecuador 
also on steep slopes where there were 
scattered bushes about 2 meters tall, and 
in eucalyptus plantations by pastures and 
streams." 

Todd and Carriker (1922) also found 
the rare X. pernix at the edge of forest, 
and in "shrubbery." A single Xolmis (ap- 
parently related to X. fusconifa and to X. 
pernix, see below) was seen by my wife 
and me on Volcan Cotopaxi in dense 
brushy paramo up to about 2V2 meters tall 



in a steep-sided quebrada which extends 
far up into the grassy paramo at about 
4000 meters. The habitat of X. erytliroptjgia 
is given by de Schauensee ( 1964 ) as 
"scrubby open slope" in paramo. 

From all accounts, the habitat prefer- 
ences of this species-group seem to be like 
those of other Xolmis and the higher alti- 
tude species of Ochthoeca. 

(3) Agriornis. Agriornis livida at Los 
Molles was in what was perhaps a natural 
habitat before the introduction of domesti- 
cated grazing animals to Chile. This sub- 
arid countryside had large patches of 
bushes, cacti, and bromeliads. Between the 
patches were relatively small but usually 
continuous areas of sparse, short, dry grass 
about 4 to 7 centimeters tall. In the same 
region the species was also on edges of 
relatively open, grazed fields with smaller 
patches of bushes and cacti not more than 
3 to 10 meters apart. Perhaps the present 
scarcity of A. livida in Chile is due to the 
opening of the environment. 

On the pass of Infieniillo in Tucuman 
Province, Argentina, the closely related A. 
montana was in arroyos at about 2200 to 
3000 meters (I did not search higher). In 
these were small streams running amid 
grass, scattered rocks, and a few bushes. 
Open, grazed grasslands separated the 
arroyos, dividing pairs from one another. 
These grasslands had few rocks and thus 
lacked elevated perches. In Chile, Goodall 
et al. (1957) report this species on rela- 
tively barren but rocky Andean slopes, and 
X'uilleumier has found it mostly in rocky 
areas. On Mt. Pichincha in Ecuador we 
found A. montana in a very open habitat at 
the lower, shallow end of a large quebrada. 
There was no grass, and the one indi\'idual 
ran on bare, dry soil much like a Miisci- 
saxicola species; it also perched and peered 
from the tops of scattered agaves and 
bushes. 

Other Agriornis (probably also montana, 
see below) individuals were found higher 
in the same quebrada on Mt. Pichincha, 
and in another, higher quebrada. In both 



Evolution of Ground Tyrants • Smith and Vuilleumier 239 



sites the slopes were well vegetated with 
bushes, bromeliads, agaves, and grasses, 
and appeared similar to the coastal A. 
livida habitat. Similar birds were found in 
even lusher habitat at about 4000 meters on 
the northwest slopes of Volcan Cotopaxi 
amid tall, dense grasses. 

On the whole, the habitat preferences 
of this genus probably fall between those 
of northern lowland Xohnis species and the 
larger Muscisaxicolo. 

(4) Ochfhoeco. Some dark-colored, rela- 
tively low altitude members of this Andean 
genus (e. g., O. cinnamomeiventris, Vuil- 
leumier's observations) live in dense 
forests, others (e.g., O. diadema and O. 
nifipectoralis) in second growth or edge 
habitats at intermediate elevations, and 
species such as O. fumicoJor primarily in 
the transition between brushy and grassy 
paramo. The relatively p)ale O. leucophrys 
inhabits arid, brushy, stream valleys over 
a wide range of altitudes. Most species 
apparently do not depend on open habitat 
and usually forage within and among trees 
or bushes. 

Locomotion and Foraging Behavior 

(1) Flight and Aerial Foraging. While fly- 
ing at least several meters, a Muscisaxicolo 
alternately flaps and folds the wings, using 
just one or very few flaps before each very 
brief fold. The flight is strong and rapid, 
and not undulating. In all but the longest 
flights the birds do not usually ascend far 
above the ground. Xohnis species have a 
similar flight, although they usually begin 
with a downward plunge from their higher 
perches and continue close to the ground. 
In both genera the end of a flight is usually 
a rapid glide on folded wings, the bird 
turning abiiiptly upward on opened wings 
to land. In short flights, birds of either genus 
may flap without intennittent folding, but 
still use the terminal glide. 

The flight of Agriornis is similar, with 
fewer wing foldings. Members of this 
genus look relatively heavy-headed and 



short-winged in flight, more thrushlike 
than the previous species. 

Both Ochthoeca rufipectoralis and O. 
fumicolor have slightly undulating flights 
with bursts of rapid beats and brief wing 
folding. In at least O. nifipectoralis a 
conspicuous pump of the tail came at the 
beginning of each burst of flapping, and 
occasionally a whirring sound came during 
a burst. In O. fumicolor whirring was 
much louder in intraspecific fights and in 
one probably interspecific situation, but 
was not heard under other circumstances. 

All of the species do some aerial fly- 
catching, and when perched, survey the air. 
In most cases the choice of foraging 
method appears to be based on availability 
of prey, but some species seem more prone 
to aerial flycatching than others. For 
the very terrestrial Muscisaxicola species 
most pursuit flights are steeply upward, 
but some extend laterally 20 to 35 meters, 
as in much more aerial genera such as 
Ttjrannus. I have seen Muscisaxicola 
macidirostris, M. bremcauda, and Och- 
thoeca fumicolor flight glean (hover in 
front of a twig or flower head and pluck 
nonflying prey). 

(2) Perching and Locomotion and Forag- 
ing on the Ground. In country without 
woody vegetation, the species of Muscisaxi- 
cola customarily perch stiffly upright on 
clods of earth, stones, or larger rocks. 
Where bushes or even trees are present, 
these provide additional vantage points. 
After scanning briefly from one point, most 
individuals dash to another, usually within 
3 meters. These rapid dashes may employ 
running, hopping, or both. Running seems 
common only if a bird goes at least a meter, 
and only on quite level, unbroken sub- 
strate. Larger species may run much more 
often than the smaller species, and, also in 
keeping with the findings of Kunkel (1962) 
for other passerines, all species seem to 
use both modes. Even a running dash may 
begin and end with rapid hopping. I have 
confirmed some observations by tracking 



240 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



the birds in licfht snow or fine earth, show- 
ing that even extremely rapid dashes may 
employ only hopping. Some species (e.g., 
M. rufivertex) often make a short flight 
instead of a dash, while others (e.g., M. 
jhvinucha) almost always dash. 

The relati\ely long tarsi of these species 
have been thought to permit greater speed 
in rimning. That running often gives way 
to hopping on broken substrates, however, 
suggests that long tarsi are probably also 
important in making long hops. Pettingill 
(personal communication to Vuilleumier) 
comments that long legs permit M. mac- 
loviono to stand high and thus see farther 
above its grassy habitat in the Falklands. 

When a Musclsaxicola species sees prey, 
it approaches by a quick flight or dash. 
Prey was often discovered at the base of 
plants where litter had gathered. Individu- 
als of at least M. flavinucha and M. front- 
alis sometimes proceeded from plant to 
plant, tossing litter aside with vigorous 
sideways swipes of the bill and peering at 
the site after each swipe. Feet and legs 
were never used for throwing litter. 

In M. olhilora and M. maculirostris, in- 
dixiduals have been seen to flash their 
wings open as they leaned forward after a 
dash or a flight toward prey. Sometimes 
this appeared to prevent them from falling 
forward by helping to check their speed. 
At other times, particularly if the bird did 
not lean forward, a very slight wing flick 
looked like an intention movement of flight 
(seen in M. albiloro, M. moculirostris, M. 
frontalis, and M. rufivertex, and by Vuil- 
leumier in M. macloviana). At yet other 
times, however, the movement in M. macu- 
lirostris appeared to be neither balancing 
nor intention. It was then ver)^ conspicu- 
ous, and might be repeated two or three 
times while standing still after a dash. In 
these cases it may have been stereotyped 
for conspicuousness, and probably func- 
tionally convergent with the "wing flash- 
ing" of other ground foraging birds like 
the mockingbirds (e.g., Mimus polyglottos, 
with conspicuous white wing markings, 



see Selander and Hunter, 1960; or Mimus 
gilvus, which lacks conspicuous wing mark- 
ings, see Haverschmidt, 1953). Convergent 
evolution of wing flashing techniques has 
been shown by Cade (1962) for the North- 
ern Shrike, Lanius excuhitor, which some- 
times forages on the ground, and similar 
techniques should perhaps be expected in 
other ground foraging birds — with or with- 
out conspicuous wing markings. Hailman 
(1960) has reached a similar conclusion, 
which he does not restrict to species that 
forage on the ground. 

Muscisaxicola species frequently use in- 
tention movements. Most (perhaps espe- 
cially M. flavinucha or M. frontalis) may 
"scissor" the tail open and closed on land- 
ing, or before and/or at the end of a dash. 
At least M. frontalis rarely gives a very 
brief, small amplitude wing flick with the 
tail movement. The frequency of tail-scis- 
soring varies greatly, even in one individual 
over a few minutes, being most common if 
the bird appears nei"vous, or if other birds 
are foraging close by. As it often exposes 
otherwise inconspicuous white outer webs 
of the outermost rectrices, it makes a good 
flock signal (see Moynihan, 1960), and 
may occur primarily during flocking. 

One M. rufivertex often flicked its half- 
fanned tail up-down at the end of a dash 
while foraging alone, usually with a very 
brief, small amplitude flick of its wings. 
Wing-flicking, not correlated with tail 
movements, \\'as seen in a foraging M. 
moculirostris. 

Each Xolmis species observed spent 
much less time on the ground than did 
species of Muscisaxicola. As Hudson 
(1920) says, these birds are "ground- 
gazers"; they usually perch on tall ^'antage 
points and peer downward, dropping to the 
ground if they see prey, or changing perch. 
Wetmore's ( 1926 ) observations of at least 
X. cinerea and X. irupero agree. I have 
seen X. irupero hover above the ground 
where no bushes were available for perch- 



ing. 



Most species hunt on the ground at least 



Evolution of Ground Tyrants • Stnith and Vuillewnier 241 



occasionally, though. I have seen both X. 
pi/rope and X. velata do this, nearly always 
hopping slowly. Only one pair of X. 
pyiope dashed and paused in the fashion 
of MuscisoxicoJa. According to Hudson, 
wintering X. rubetra and Neoxolmis rufi- 
ventris run on the ground much more than 
the other species. J. Delius (personal com- 
munication) is familiar with the latter in 
winter and says that it does not make the 
repeated short dashes typical of Muscisaxi- 
cola species. Wetmore (1926: 302), how- 
ever, describes X. rubetra as running 
"swiftly along the ground to pause and 
stand with head erect." 

I have never seen XoJmis species take 
prey that was difficult to kill, but Hudson 
( 1 920 ) says they typically grasp beetles or 
grasshoppers with their feet while strug- 
gling to kill them. 

Vuilleumier has seen X. sfriaticoJlis fly 
down from perches and forage in 30-centi- 
meters tall grass. Little is known of the 
habits of this or other members of the 
fumigata species-group, but from his ob- 
servations they appear similar to other 
species of Xohnis or to Agriornis, as might 
be expected from their size and structure. 

Foraging of Agriornis livida and A. 
montana is similar to that of the previous 
genera, and in some ways intermediate. 
Periods of a minute or more are spent peer- 
ing about from the top of a bush, brome- 
liad, or a rock, followed by dropping and 
running to another perch. Sometimes food 
is taken from the ground during such a 
dash; this has been large invertebrates in 
the few cases I have seen. At least A. 
livida sometimes gleans food from branches 
accessible from its perch. Goodall et al. 
( 1957 ) say these species take large prey 
(small mammals, lizards, amphibians, and 
birds' eggs and nestlings), as well as in- 
sects. Goodall's figure of A. livida shows a 
bird holding a mouse under one of its large 
feet. Peiia ( 1961 ) recovered seeds, plants, 
insect remains, and a wing-bone of a small 
bird from stomachs of A. montana; he also 
observed an A. montana killing a "small 



mouse." Birds of this genus, even more so 
than those of Xohnis, quite likely grasp 
prey in their feet and beat it with their 
bills. They appear to be, as Goodall et al. 
claim, ecological counterparts of shrikes. 

The diree Ochthoeca species I have seen 
all often pumped the tail (up-down, as in 
all of the species in this paper) with a 
simultaneous brief, small amplitude flick 
of the wings. These movements came on 
landing or when perched, and their fre- 
quency increased if I made a bird nervous. 
O. leucophrijs and O. fumicolor both did 
mostly aerial foraging, and the latter 
several times sallied out to flight glean 
from the paramo grass or to take prey 
flying over the paramo. In 1959, while an 
O. fumicolor kept in the edge of a grove 
or worked over the paramo, an O. rufi- 
pectoralis worked through the same grove, 
hopping from twig to twig and gleaning 
like a Vireo. Vuilleumier has seen O. leuco- 
phrijs fly down to the ground for prey. 

Although the data are scanty, it appears 
that no known species of these genera is 
primarily an aerial flycatcher. All spend 
considerable time as gleaners, taking much 
of their food from the surface of the ground 
or of the vegetation. 

Flocking 

In at least central Chile, while lush early 
spring sites are available and snow-covered 
higher slopes are not, large numbers of 
Muscisaxicola flycatchers may forage to- 
gether. The species of large flocks inter- 
mingle, to some extent, with less abundant 
species clumping amid the more numerous 
individuals of other species. I have never 
seen a fully stable flock, and there is al- 
ways some intra- and interspecies aggres- 
sion. Birds of larger species often supplant 
individuals of smaller species; individuals 
of different species that are about the same 
size occasionally fight vigorously. 

As higher sites open and the flocks thin 
out, aggression appears to increase. I have 
then seen temporary territories in which 
one individual foraged for some minutes 



242 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



and drov'e out all intruding conspecific 
individuals. Individuals of other species 
might be driven out or tolerated. These 
appeared to be foraging territories only, 
and were small — three to seven meters 
across in most cases — in sites that were 
exceptionally wet and covered with low, 
green vegetation. 

Hudson (1920) mentions single-species 
flocks of several wintering species of 
Xolmis in Argentina. Further, Neoxohuis 
rufiventris is said by him to associate with 
flocks of plovers. Whether these were 
really mixed flocks or aggregations result- 
ing from patchy distribiition of habitat is 
not clear. The same may be said of the 
interspecies flocks of Muscisaxicola. 

I have found no indication that members 
of the other genera (Agriornis and Ocli- 
thoeca) flock outside the breeding season. 
Within the breeding season all are reputed 
to be found only in pairs. Flocking be- 
havior does not seem to be characteristic 
of this group outside of migration, and may 
play an important role only in Xolmis. 

Nest Construction 

The open, cup-shaped nests of Musci- 
saxicola species are always placed on the 
ground in crevices between rocks, or even 
within a rabbit burrow (e.g., M. capistrata, 
personal observation). Only M. brevi- 
cauda does not always employ crevices, 
and where its nests are hidden by grass 
alone they are often partially roofed with 
grasses (Marchant, 1960). 

Most Xolmis species build open cup- 
shaped nests in bushes or trees, although 
X. iriipero hides its in holes in trees (Hud- 
son, 1920; Wetmore, 1926), or appropriates 
an enclosed nest of Fnrnarins rufus (see 
Hudson, 1920). X. cinerea is said to nest 
in holes in banks in Argentina, but to build 
an open nest in trees in Brazil and Uruguay 
(von Ihering, 1904; Hudson, 1920). X. 
coronata nests either in a bush or in a 
"large clump of grass" (Hudson, 1920), but 
apparently not on the ground. The only 



described nest of Neoxolmis rufiventris was 
on the ground (Maclean, 1969). 

The species of Agriornis also build cup- 
shaped nests in bushes, except for A. mon- 
tana, which nests on the ground amid 
rocks like a Muscisaxicola (Goodall et al., 
1957). Nests of OcJitlioeca species and 
Pyrocepliahis ruljinus are open cups, usu- 
ally in bushes or trees. Todd and Carriker 
(1922) report a nest of Ochthoeca rufi- 
pccforalis that was placed on "a projection 
on the side of a huge boulder, sheltered 
by some stunted trees," apparently com- 
parable to the nests of Saijornis species. 

PART II. DISPLAY BEHAVIOR 

The following accounts present what is 
known about the form and employment of 
stereotyped signals ("displays") used by 
these birds in communicating. A detailed 
description of the display repertoire of 
Muscisaxicola alhilora provides a basis for 
comparison. The displays of two small 
Muscisaxicola species are also described in 
detail: that of M. maculirostris to show its 
comparable structure, even though it 
sounds simpler to the human ear, and that 
of M. hrevicaucla because (a) it is at least 
superficially somewhat aberrant; and {h) 
the species has customarily been placed in 
the monotypic genus Muscigralla, usually 
not seen to be related to Muscisaxicola 
(Hellmayr, 1927: 94, was not even sure it 
belonged in the Tyrannidae). 

Where possible, comparisons are made 
with displays of Sayornis species, since 
these are known in much greater detail and 
provide a coherent frame of reference (see 
Fig. 11). Occasional comparisons among 
species are made in this section to facilitate 
description, but general comparisons fol- 
low in the Comparative Summary. 

Ma. Muscisaxicola 

Muscisaxicola albilora 

Most individuals of this species were 
still in loose flocks, although many ap- 
peared paired, on 5 November 1961, at 



Evolution of Ground Tyrants • Smith and Vnillcumier 243 



Farellones and La Parva, Chile. On 7 and 
8 November those at the Rio Yeso sites all 
appeared paired, but were foraging com- 
munally at the best sites and showing only 
slightly more frequent territorial behavior. 
Several kilometers farther into the interior 
of the Andes along the Rio Volcan on 8 
November there were flocks of over 60 
individuals, plus lesser numbers of at least 
two other congeneric species. By 17 No- 
vember most snow at the upper Rio Yeso 
site had melted, and much more green 
vegetation was scattered about. Most pairs 
had dispersed onto the slope, and I found 
one nest being built among the boulders on 
an unstable talus slope. There was some 
communal foraging at the best local site, 
however, the edge of a large meltwater 
pond. 

In 1962 the upper Rio Yeso site was al- 
most entirely snow-covered on 15 October, 
and had no M. olhilora. On this and the 
subsequent day only three small groups 
were at the lower site; the individuals as- 
sociated with almost no display and only 
occasional supplanting attacks. A few lone 
individuals occasionally held temporary 
territories, but \\'ith little display. 

Vocal Displays 

(1) In agonistic encounters increasingly 
prominent aggressive tendencies are ex- 
pressed by the more or less graded series: 
"tseet" and "tseek," "tut," "tsk," "tchk," and 
"tk." Most are known primarily from males. 
Both "tseet" and "tseek" (Fig. la and b) 
may occur very early in agonistic en- 
counters, and "tut" (Fig. 2c) and "tsk" 
(Fig. Id) at various points, but "tchk" (a 
more sharply peaked variant of "tsk") and 
"tk" (Fig. 2d, perhaps a high, brief, ex- 
treme variant of "tut" with different har- 
monic structure, an extended peak, and 
prominent descending terminal arm) are 
almost restricted to fighting, as two birds 
grapple. 

(2) As avoidance or escape become 
more probable, "tseet" or "tsk" calls are 



replaced by more prolonged vocalizations: 
"teek" (Fig. Ic) and "sect" (Fig. le, and 
with a slight tendency to show two peaks 
in Fig. If), respectively. During agonistic 
encounters, fights are often separated by 
periods in which the quarreling birds stand 
apart and usually do Wing Raising displays 
while one individual utters long strings of 
"seet." In at least some encounters, how- 
ever, it was this individual which later 
pressed the attack again. 

(3) A bird flying or landing alone may 
utter a single "tseet." This is more likely 
to be a "tseek" or especially a "teek" (Fig. 
Ic) if the commimicator is flying toward 
its mate. The usual reply by the mate is a 
"tsk," but if both individuals are in flight 
toward each other there is often a rapid 
jumble of quite variable calls (e.g.. Fig. 
Ig), perhaps not all uttered by one indi- 
vidual. These include "tseek," "seet," "tsk," 
and even "tchk," plus special forms such as 
the third and final in the illustration, which 
are probably the only ones peculiar to 
greeting usage. Sometimes in less agonistic 
situations, however, mates employ a simpler 
stereot\'ped greeting (Fig. Ih). 

Bursts of a simple, chevron-shaped call, 
"tuh" (Fig. 2a), have been heard and re- 
corded only from apparently paired birds 
foraging together when there has been no 
agonistic activity for at least several 
minutes. Bursts of a higher form (Fig. 2b) 
appear intermediate to some greetings. The 
similar "tut" call (Fig. 2c) is sometimes 
uttered, singlv, from mates while foraging. 

(4) In one Aerial Display a loud 
"clee-ip" (not recorded) was heard; it was 
quite reminiscent of the corresponding call 
of M. maciilirostris, but with a more abrupt 
termination. 

Most of the known vocal displays of M. 
alhilora are thus unelaborate calls linked 
by extensive sets of intermediates. Only 
the most different forms, such as "tuh" and 
"tseet" are easilv' distinguished by the 
human ear, although many of the others 
are recognizable with practice. The more 
agonistic vocalizations resemble in form 



244 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



■ 






A 






. a 


b 


c 


d 


e 
■ 1 


f 



■ 












A 


^ 


^ 


A 


'\ 


A 


^ 


m 


A 


A 


A 


^ 


a 






' 




> 



u 



'i 






^i A 



^ ^ 



^ 



-1 — 1— 




Figure 1. Vocalizations of Muscisaxicola albilora-. la. 
"tseet"; lb. "tseek"; Ic. "teek"; Id. "tsk" (compare with 
Fig. 11a); le. "seet '; If. "seet" (note tfie slight tendency 
toward two peaks in this case); Ig. an example of a variable 
greeting; Ih. an example of a greeting in a less agonistic 
situation. 



A 



Figure 2. Vocalizations of Muscisoxicolo albilora-. 7a. a 
burst of "tuh" calls; 2b. a burst of higher "tuh" calls (note 
general similarity to Fig. Ig); 2c. "tut"; 2d. two examples 
of "tk." 



and usage calls of Soyornis species. For 
instance, "tseet" is comparable to the Initi- 
ally Peaked Vocalization (IPV) of all 
Sayornls species, "teek" resembles the 
Simple Vocalization (SV) of S. nigricans 
and "tsk" the SV of S. plioebe, and "tseek" 
somewhat resembles the Bipcaked Vocali- 
zation (BV) of S. phoehe and S. nigricans. 
As in at least S. phoelie, greetings between 
mates during or shortly after agonistic situ- 
ations usually involve SVs, BVs, or IPVs, 
and the special greeting vocalization of M. 
albilora resembles that of S. saya and to a 
lesser extent, that of S. phoehe. 

None of these possibly homologous calls 
corresponds exactly to each other in spe- 
cific usage. The fit is probably least good 
for the bursts of "tuh" calls of M. albilora, 
which may be only superficially similar to 
the Chatter Vocalization (CV) of S. saya, 



and the Doubled Vocalization (DV) of S. 
phoehe. (In comparing forms, note that 
the scale of Figure 11, showing sample 
Sayornis vocalizations, is like that used for 
the remaining genera of this paper, but 
proportionately condensed along the time 
axis and elongated along the frequency 
axis as compared with the scale used to 
illustrate the very brief Muscisaxicola 
vocalizations.) 

Nonvocal Displays 

(1) Wing Raising. In prolonged aggres- 
sive encounters birds within a meter of one 
another may raise one or both \\'ings. En- 
counters in which neither individual is \'ery 
actix'e usually lack Wing Raising. The 
minimal form has been seen when two 
birds, sometimes one member of each of 
two pairs, approach each other during 



Evolution of Ground Tyrants • Smith and Vuillemnier 245 




Figure 3. Some Visible Displays of Muscisox/co/o alb/loro: 3a. minimal Wing Raising with one wing, and slight Head 
Bow; 3b. Wing Raising with synchronous fluttering of both wings; 3c. asynchronous Wing Raising with Crown Ruffled, and 
ruffling of the flanks and chest. 



foraging. One or both may "casually" lift 
one wing very slightly (Fig. 3a) without 
other display, and not facing its opponent. 
In such instances no fights result, and the 
birds usually go their separate ways. 

There may be fights, however, when two 
or more pairs come together after prebreed- 
ing flocks have scattered. Simple raising 
of one wing by one or more individuals 
comes early in such encounters; later the 
movement becomes jerky and abrupt, and 
the second wing begins to be jerked slightly 
up. Often the two wings are lifted to 
different extents and asynchronously, or 
both are raised at once and wagged ir- 
regularly in an asynchronous, "semaphor- 
ing" pattern emphasized by the pale color 
of the wing undersurfaces. Less commonly, 
both are lifted high over the back and 
fluttered in unison (Fig. 3b). Rarely, a 
bird in this pose flies perhaps three to five 
meters between two rocks, still with its 
wings high and fluttering. This resembles 
part of the Aerial Display, and it is likely 
that Wing Raising and that display inter- 
grade, although I have not seen the re- 
maining intermediates. Any flight during 
a Wing Raising encounter will likely ter- 



minate with the wings held high, and 
partial raising in what appears to be a 
flight intention movement is common be- 
tween bouts of Wing Raising. The tail is 
fully spread in most forms of the display, 
but not flicked. 

The flanks and chest are ruffled during 
at least some phases, and the Crown 
Ruffled accompanies all but the simplest 
forms (Fig. 3c). Thus Wing Raising shows 
several special plumage features: pale wing 
linings, pale bases to the primaries on the 
upper wing surface, and the occipital 
crown-patch. 

Opposing individuals orient variously but 
usually laterally; a tendency to orient side- 
ways to the opponent seemed strongest 
when both wings were held high. I could 
detect no obvious preference to raise, or to 
raise first the wing closer to or farther 
from the opponent. 

A few times, up to five individuals, ap- 
parently including t\\'o pairs, partook in an 
encounter. In the most ^'igorous periods of 
activity all used Wing Raising. Rapid and 
confused chasing and fighting developed, 
and long chases went beyond effective 
observing range. In one encounter between 



246 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



two pairs tlie display of two individuals 
proceeded as far as two-wing semaphoring 
with only an occasional "tseet," and no 
fight or chase resulted. After about a 
minute of stalemate one individual per- 
formed an Aerial Display. 

Sayornis soya and probably S. nigricans 
have displays similar in form to Wing 
Raising and the forms of Aerial Display 
closest to it, but these appear to function 
in what is probably pair-bonding. As I 
have seen neither territorial defense in 
those two Sayornis species, nor pair-bond- 
ing in M. alhilora, the functional compari- 
son is incomplete. 

(2) Aerial Display. On 5 November 
1961, an individual several times silently 
chased an intruder from what was probably 
a potential breeding territory. It then be- 
gan to fly, climbing slowly with wings held 
high over its back and used in bursts of 
short strokes. Its long legs dangled and the 
feet appeared relaxed. The dark tail was 
held vertically down and fully spread. The 
bird remained silent. It often simply 
hovered, sometimes climbed higher, and 
frequently turned to a new direction, re- 
maining over a limited area. Finally it 
descended in a long, fast swoop with its 
wings closed until near landing, and 
perched stiffly upright atop a group of 
rocks. 

At another site on 17 November 1961, a 
bird left a stalemated Wing Raising en- 
counter (q.v. ) and flew up in a Tyrannus- 
like twisting flight of about 20 to 25 meters, 
then swung around and dropped c|uickly 
to a foraging area. During the flight it 
uttered several loud, abrupt "clee-ip" calls. 

(3) Crown Ruffled. During Wing Rais- 
ing disputes all birds ruffled their crowns, 
either entirely or just at the rear. This 
made the head appear larger, and also 
made very conspicuous the occipital 
crown-patch, the rusty color of which is 
species-specific. 

Establishment of a temporary foraging 
territorv was watched on 16 October 1962. 



The owner foraged vigorously for several 
minutes before a conspecific individual 
arrived. The intruder was immediately 
driven to the edge of the area without dis- 
play, but it persisted in reintruding over 
the next three minutes. The territorial 
owner always caused it to retreat by dash- 
ing toward it without display. The tres- 
passer never faced toward the owner, and 
^^'ould ruffle the back of its crown as soon 
as it began to be pressed and on stopping 
after a chase. 

(4) Head Bow. On 5 November 1961, 
two individuals were observed about 60 
centimeters apart, oriented almost toward 
one another and bowing their heads, thus 
revealing the ruffled patch of color at the 
back of the crown. Occasionally one or 
both uttered a barely audible "tseek." Re- 
fore anything further happened one saw 
me and ran off, the other following several 
meters behind. 

SAuscisax'fcola frontalis 

In 1961 individuals of this species were 
observed foraging by a pond where some 
interacted with M. alhilora. In 1962 three 
were observed foraging in loose association. 

Actively foraging individuals uttered two 
different calls: "trit" and "treet" (Fig. 4a 
and 4b). They resemble, but at much 
lower frequencies, "tseet" and "sect," re- 
spectively, of M. alhilora. One individual, 
apparently nest site prospecting while 
foraging, uttered short bursts comprising 
several "pt" vocalizations at two frequen- 
cies (Fig. 4d); these were like the "tuh" of 
M. alhilora, and the usage strengthens the 
comparison of such calls with the CV of 
Sayornis. Yet another foraging individual 
used a "tuk" (Fig. 4c, like the "pt" but 
higher, briefer, and uttered singly instead 
of in bursts — i.e., resembling the "tut" of 
M. alhilora). 

I recorded several sharp "t" calls from 
an M. frontalis as, and shortly after, it 
fought with an M. alhilora. These resemble 
the "tk" used by the latter species in fights, 



Evolution of Ground Tyrants • Smith and Vuilleumier 247 



but emphasize the initial ascending arm 
and ahnost omit the descending arm. All 
known vocalizations of M. frontalis thus 
correspond in form, and in general usage, 
to calls of M. albilora. 



\ 



Muscisaxicola flavinucha 

Displays were observed only on 26 Sep- 
tember 1962, when a heavy spring snowfall 
at Lagunillas, Chile, brought birds down 
from higher altitudes. Three M. flavinucha 
appeared on my study area before snow 
accumulation exceeded 2.5 centimeters. 
They associated loosely while foraging, 
and one made occasional silent supplanting 
flights against the others. 

Once the aggressor stopped about 60 
centimeters away from another without 
supplanting. The approached individual 
stood still with its head withdrawn be- 
tween its shoulders, and began to call. 
There were some short bursts such as "tsee 
tee tsee tseet." Other calls were more ir- 
regularly spaced, but sounded similar, and 
were like calls recorded from M. albilora 
during greetings. Both ruffled their upper 
backs. The calling one maintained an 
oblique orientation toward the other for 
about 90 seconds, then the latter wandered 
off, and eventually flew away. 

Vuilleumier ( personal communication ) 
has heard vocalizations from two adults 
feeding large nestlings at Cerro Llaima 
(Cautin Province), Chile: high-pitched 
"seeht" and "seeseeht." A fledgling he 
watched being fed at Cerro Catedral (Rio 
Negro Province), Argentina, begged with 
a thin "see .... see . . . ." 

SAuschaxicola rufivertex 

On 12 October 1961, several separate 
individuals were seen on steep slopes near 
Lagunillas, Chile. In mid-aftemoon two 
flew from opposite sides of a rocky stream 
bed out over a gully, somewhat toward one 
another without approaching closely. Both 
performed Aerial Displays. Each would fly 
about 16 meters then rear up until hanging 



Figure 4. Vocalizations of Muscisaxicola frontalis-. 4a. 
"trit" (compare with Fig. la); 4b. "treat" (compare with 
Fig. If); 4c. "tuk " (compare with Fig. 2c); 46. burst of "pt" 
vocalizations (compare with Fig. 2a and 2b). 

vertically and stall with wings stretched 
fully above its back, at this point uttering 
a thin, high-pitched "twee-it." This was 
repeated several times, then both turned 
back and repeated it several more times 
before landing on the slopes. Neither dis- 
played further, and later in the season they 
apparently ascended to higher altitudes as 
the snows melted. 

On 16 October 1962, two apparent pairs 
foraged on the rocky outu'ash plain of a 
small stream at the lower Rio Yeso site. 
The one pair flew to a bush about 1 meter 
tall, one perching on top and aligning at 
right angles to the pair still on the ground. 
This upper one began occasionally lifting 
and waving one wing, the wing toward 
the other pair. It had its chestnut occipital 
patch conspicuously elevated (i.e.. Crown 
Ruffled). The other pair was then about 
10 meters away, the nearer individual fac- 
ing the bush and Wing Raising vigorously, 
usually with both wings. Within a minute 
they flew toward the same bush, landing 
within 60 centimeters of the first pair. The 
first indi\'idual now aligned slightly more 
away, but continued to raise its wing. The 
"aggressor" oriented its body to within five 
to ten degrees of the first and raised both 
wings. Both had Crowns Ruffled, and I 
could hear "tee" vocalizations poorly over 
the high wind. 

Within a minute, the individual which 
had approached dropped back to a second 



248 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



V 



( . 'i 




Figure 5. Vocalizations of Muscisaxicola caphtrata-. 5a. 
"tueee" (compare with Fig. la); 5b. "tee" (compare with 
Fig. le and If); 5c. a common vocalization, intermediate be- 
tween the previous two; 5d. clear "seeee ' of a nestling, 
begging; 5e. loud and harsh call of a nestling, in the hand. 



bush 70 centimeters farther away, and 
continued visiorous Wine; Raisins;. The first 
then flew about 6 meters followed by its 
apparent mate; the other pair followed to 
within 2 meters, and all displayed as be- 
fore. The first pair then flew another 6 
meters away, the displaying individual 
raising both wings over its back and glid- 
ing to its perch. This ended the encounter; 
within 45 minutes all four birds had left 
the area. 

All known displays of this species are 
comparable in form and usage to displays 
of M. olbilora. Note that Wing Raising 
\\'ith both wings characterized the more 
aggressive bird, and wings raised high in 
flight appeared in a situation similar to 
that in which it occurred in M. albilora, 
again resembling an intermediate between 
the Aerial Display and Wing Raising. 

Muscisaxicola macloviana 

Hudson ( 1920 ) described "low plaintive 
whistling notes" given by this species when 
wintering, perhaps what Pettingill (per- 
sonal communication to Vuilleumier ) called 
a "reedy cheep." 

Muscisaxicola capistrafa 

Observations were made from 4 to 8 
December 1961, on the farm "Las Mer- 



cedes," Tierra del Fuego, Chile. One nest 
was located with three young, about a 
week old. Other local pairs probably had 
nests, and there were also apparently a 
few independent fledglings. 

Vocal Displays 

(1) While I examined nestlings, one 
adult sometimes called a "tueee" (Fig. 5a), 
closely similar to the "tseet" of M. albilora, 
but witli its two peaks respectively low and 
high instead of the reverse. It did not ap- 
proach close to me and usually called only 
when its mate flew to perch near it. 

( 2 ) A variable "tee" ( Fig. 5b ) was used 
by both parents while I examined their 
nestlings, and the one which approached 
me closer uttered only this call. It is 
similar to the "sect" of M. albilora, the 
main difference corresponding to that of 
the previous case: the second peak is 
higher in M. capistrata. 

Intermediates between "tueee" and "tee" 
(e.g.. Fig. 5c) were common on each oc- 
casion I examined the young. During five 
consecutive days, however, I heard no 
other vocalizations from any adults. 

(3) Nestlings removed from the nest- 
burrow uttered two quite variable calls 
(Fig. 5d and e). One was a simple, clear 
"seeee" ^^ith rising inflection, and the 
nestlings were sometimes begging when 
uttering it. It resembles the IPV of Saijor- 
nis saya nestlings. The other was loud, had 
an initial ascending arm, terminal descend- 
ing arm, and often something resembling 
a chevron between, but with gaps filled 
with harsh noise separating all three sec- 
tions. 

Nonvocal Displays 

(1) Feather Bufflinfi. The parent which 
approached me closely uttering "tee" re- 
mained conspicuously sleeked. Its mate 
usually perched 3 to 7 meters away, called 
"tee" frequently, and sometimes ruffled its 
chest, flanks, and the sides of its head. 
(At least the first two areas were ruffled 



Evolution of Ground Tyrants • Smith and Vuilleumier 249 



by some M. albilora when Wing Raising, 
but the present species did not Wing Raise 
in the several intraspecific fights observed. ) 

(2) Wing Wliin: Very loud Wing Whir- 
ring was heard in some intraspecific fights, 
and was likely a display. Loud Wing 
Whirrs were also produced by the bird 
hovering near me at the nest. (A Wing 
Whirr was also heard from one M. frontalis 
in a situation I could not be sure was dis- 
play, but possibly this sound need never be 
produced in any wing stroke functioning 
solely for flight.) Its usage corresponds to 
that of a similar sound of Sayornis phoehe. 

(3) Aerial Display. On the afternoon of 
4 December a lone individual made two 
silent display flights within three or four 
minutes. The bird made little headway 
against a wind of about 30 kilometers per 
hour. It simply flew up about a meter, 
raised its wings high over its back, and 
fluttered briefly until poised in a vertical 
stall with wings stretched maximally up- 
\\'ard. Then it fluttered again, repeating 
the procedure two or three times per flight. 
This was probably the sort of intennediate 
between Wing Raising and full Aerial Dis- 
play which was missed in M. aJhilora. 

Muscisaxicola maculirosfris 

I observed this small species during 
several periods: shortly after its arrival on 
breeding grounds (13 September to 16 
October 1962, Chile), during nesting (9 to 
17 June 1966. Ecuador, and 28 November 
1962, Tucuman, Argentina), late in the 
breeding season (4 to 6 August 1959, Ecua- 
dor ) , after the breeding season ( 13 Janu- 
ary 1962, Chile), and in midwinter (4 to 
6 February 1962, Ecuador). Although at 
one Chilean site (Lagunillas) and at all 
Ecuadorean sites near San Antonio popu- 
lation densities were high, there was never 
much interaction or much displaying. 

Vocal Displays 

( 1 ) Its basic call is "tek," most variants 
of which are indistinguishable in the field. 
Extensive tape recorded samples show that 



its variations tend to correspond to the 
various briefer calls of M. albilora, but are 
insufficient to show the limits of each. 

The "tek" itself probably corresponds 
roughly to the "tut" form of M. albilora, is 
often repeated at similar intervals, and has 
the same chevron shape but not the har- 
monics (Fig. 6a). Foraging individuals 
often repeat it for minutes at a time, but 
are silent for similar periods. Usually it 
comes at or just after the end of a dash, or 
while the bird is stopped and peering 
around. 

A bird which has just been fighting, or 
which has watched a hawk or owl pass by 
may utter "tek" and slight variations 
frequently. It was occasionally used in 
mobbing a small predator such as the 
Burrowing Owl (Speofyto cunicularia) or 
Sparrow Hawk (Falco sparverius), at least 
in Ecuador \\'here these are common and 
were often ignored or mobbed in a "des- 
ultoiy" fashion without displays. Some- 
times when we appeared to make an 
individual slightly nervous it would begin 
to call "tek." 

Occasionally, slightly higher pitched 
forms can be distinguished. These are most 
often uttered by a bird about to join an- 
other (or about to be joined), or one 
which is foraging near another and appears 
slightly nervous. They probably cor- 
respond to the "teek" of M. albilora. Other 
variants are lower or more prolonged, or 
have a more prominent tail (like a small 
"whee-oo" call, q.v. ). The prolonged ones 
may be used primarily by relatively fearful 
birds. No known vocalization clearly cor- 
responds to either the bi-peaked "tseek" or 
the "tseet" of M. albilora. 

(2) I have recorded nothing as flat as 
the prolonged "sect" of M. albilora, but did 
obtain some relatively long, variable, asym- 
metrical, squeak)^ calls (e.g.. Fig. 6b) from 
an apparently partly cowed and indecisive 
individual. It had begun to crouch after 
running toward several indixiduals that 
were responding to the overflight of a large 
hawk with Aerial Displays. The area was 



250 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



the site of several intraspecific encounters 
and was likely a boundary region among 
ill-defined territories. Once a grappling 
fight there terminated on the ground, then 
the attacked bird crouched low with wings 
outspread and uttered a "squeal" ( probably 
more prolonged than in Fig. 6b) in the 
brief interval before it was again attacked. 

(3) No greetings were recorded from 
mates meeting in agonistic circumstances. 
However, two probably intergrading sorts 
of calls were found, each with a single peak 
preceded by a shaiply descending arm. 
The forms resemble BV-types recorded in 
agonistic greetings of M. alhilora. The first 
(Fig. 6c) is relatively prolonged, and was 
recorded from the individual which next 
uttered the call shown in Figure 6b, just 
before it stopped approaching the others. 
The second (Fig. 6d) tends to occur in 
agonistic encounters, including fights, and 
may be repeated in series before an Aerial 
Display (Fig. 6g shows the end of such a 
series). Such series, however, usually be- 
comes composed just of "tk." 

In relatively nonagonistic meetings mates 
seldom called, and then usually used "tek" 
or a closely similar SV variant. I very rarely 
heard calls that could have been similar 
to the LHV-type display of M. alhilora, 
and obtained no recordings. 

(4) The "tk" is slightly briefer and lower 
than "tek." It may be uttered in series 
before launching an Aerial Display, or in 
rapid series while standing and watching a 
neighbor's Aerial Display. Other uses cor- 
respond to the known uses of "tuh" series 
by M. alhilora. 

The descending arm of "tk" is character- 
istically more emphasized than is the 
ascending. From two fights, one in Ecua- 
dor and one in Chile, I recorded versions 
virtually reduced to a prominent descend- 
ing arm (Fig. 6e and f), and thus compa- 
rable to its apparent homologue in M. 
alhilora. The Chilean example is much 
lower in frequency than the Ecuadorean. 
Shortly after the Chilean fight (between 



two males establishing territories), one of 
them did an Aerial Display and upon 
returning gave a similar burst of low "tk" 
as he landed alone. 

Similar brief "tk" variants, usually at 
fairly low frequencies and not in series, 
were also used in fights, and many by an 
individual standing outside one fight as a 
spectator. He became the aggressor only 
when the first aggressor withdrew. 

(5) The Aerial Display vocalization has 
two parts, the second of which commonly 
occurs alone when the display is used after 
dawn. The first is a series of "tk" or some- 
times "t," which seems to be a very brief 
"tk." This accelerates toward the end, 
then rises in frequency either through 
several units ( Fig. 6g ) or in one unit ( Fig. 
6h). The second part is a clear, whistled 
"clee-oo" or "whee-oo." In one Ecuadorean 
bird this had a relatively prolonged tail 
and harmonics (Fig. 6i). When combined, 
the two parts resemble the Regularly Re- 
pcat(xl Vocalization (RRV) of Pyrocepha- 
his ruhinus (Smith, 1967) and the CV plus 
IPV of Sayornis saya (Smith, in press, 2). 

(6) The sole nestling examined repeated 
almost continuously a faint, rapid series of 
variable "ti" units (Fig. 6j) as my wife 
held it in her hand. This is not yet known 
from congeneric species, but in Sayornis 
phoehe and S. nig,ricans the "tee" of nest- 
lings is very similar and may be used in 
bursts. 

Almost fully grown fledglings sometimes 
uttered rapid series (probably correspond- 
ing to the nestling call) and gaped toward 
an adult. When they were about to be fed 
the series seemed to become high pitched. 
If I approached such a fledgling closely it 
would usually begin to call "tek." 

Nonvocal Display 

(1) Aerial Display. In the most complex 
form the bird flies out (or outward and 
upward if the slope is not steep) and be- 
gins a flat, fluttering fhght, calling a string 
of very brief "t" or "tk" units, sometimes 



Evolution of Ground Tyrants • Smith and VuiUeumier 251 



jerking slightly back with each call. As the 
series accelerates the frequency of the last 
few units ascends and the bird climbs 
sharply into a stall. Just before stalling, the 
tail flicks up, then drops straight down- 
ward and spreads. The bird comes to hang 
vertically in the air, dangling its legs and 
feet. Its wings stretch over its back and 
nearly touch together as it utters the clear 
"whee-oo" and then pitches forward. Re- 
gaining speed, it retracts its legs and flut- 
ters on in another performance. The flight 
is straight or twisted, and in the latter cases 
remains over a small piece of ground. After 
from one to about fifteen ( commonly about 
five) stalls, the bird flies back to the slope 
or partly closes its wings, slightly cocks its 
tail, and dives to the ground. Peters (1923) 
gave an accurate but much less detailed 
description. 

There are various less complex forms. 
Just the terminal "whee-oo" may be uttered, 
which is very like counterparts of the dis- 
play known in M. albilora and M. rufi- 
vertex. At least in Ecuador in 1966, this 
was the common form after extensive 
predawn usage of the full form. Flights 
were then usually given in response to the 
passage overhead of an avian predator or 
in response to Aerial Display by a neighbor. 
The latter individual might be close to a 
boundary and the two displaying against 
one another, or (more commonly) might 
be responding to an avian predator — the 
first Mnscisaxicolo to go into Aerial Dis- 
play when a Sparrow Hawk or Buteo 
passed seemed to precipitate the display 
from most of his neighbors. 

Sometimes a bird calling "tek" would 
launch into Aerial Display without an 
introductory series. More commonly, a 
series of "tk" calls was uttered before the 
bird took flight — these may have replaced 
the usual aerial calls. Sometimes such a 
"tk" series would build up in speed and 
then slow down, the bird lapsing into "tek" 
without an Aerial Display. Fully complex 
Aerial Displays were never preceded by a 
"tk" series from the ground. 



- 






\ 






- h 




\ 




\ 


\ 


a 


b 


c 


d 


e 


,^ 



\^ 




r 





^ ^ 



^ ^ '^ 



Figure 6. Vocalizafons of Muscisaxicola maculirostrls: 6a. 
"tek" (compare with Fig. Ic, and first vocalization of Fig. 
Ig); 6b. squeaky sounding vocalization (compare with Fig. 
le and If); 6c. vocalization recorded from an individual 
running toward a fight, just before stopping and crouching; 
6d. a somewhat similar form, used in fights, agonistic en- 
counters and sometimes in series before an Aerial Display; 
6e. "tk," recorded from a fight in Ecuador (compare with 
Fig. 2c and 2d); 6f. "tk, ' recorded from a fight in Chile 
(compare with Fig. 2d); 6g. vocalizations from an Aerial Dis- 
play: the end of a "tk " series, with the last intervals short- 
ening, and the last several "tk" rising in frequency before 
the "whee-oo"; 6h. vocalizations from an Aerial Display: 
the "tk" series is not shown, but ended in a single relatively 
high unit, followed by the "whee-oo"; 6i. a "whee-oo" 
vocalization with a relatively prolonged tail; 6j. three suc- 
cessive "ti" vocalizations from a series uttered by a nestling. 



In a unique variant the terminal 
"whee-oo" was omitted and the aerial 
stuttering series left intact. The bird flew 
up when a Sparrow Hawk pounced well 
upslope from him, and did ordinary Aerial 
Display with five stalls and vocalizations 
before each. His "t" series accelerated and 
rose slightly in frequency just before stall- 



252 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



ing, but at the stall he simply used another 
"t" of the original frequency. Yet another 
individual once gave a single performance 
that was silent except for a sharp "clit" at 
the stall; shortly later he did an Aerial 
Display with a "whee-oo" at each of several 
stalls. 

The display's appearance also varies. In 
the most interesting case, an Ecuadorean 
bird rose to level with the lip of a quebrada 
then flew at that height, repeating rapidly 
"t t t tu-twit" in which the terminal arm of 
the call seemed foreshortened (not re- 
corded). With each "tu-twit" it arched its 
back, raised its wings high (but not to the 
customary extreme), lowered its legs, and 
partly cocked and partly spread its tail. 
This cocking of the tail is normally brief 
and precedes the stall. In this case, the 
bird did not assume a vertical orientation 
of its body, but used a form much more 
like that of M. brevicouda. 

The Aerial Display may not differ ap- 
preciably between the Chilean and Ecua- 
dorean populations observed. The full 
form most common in Chile was also most 
common in the preda\\Ti twilight in Ecua- 
dor. The Ecuadorean vocalizations are 
apparently slightly lower in frequency, and 
the terminal "whee-oo" at least sometimes 
descends less abruptly than in the Chilean 
sample. 

Not all factors governing the use of this 
display are yet known. It is more common 
just before and during the nesting season 
than at other times. It is "infectious" in 
that males stimulate one another to per- 
form ( as in the Tumble Flight of the genus 
Tijnmnus, Smith, 1966, and the display 
flights of Pyrocephahis ruhimis, Smith, in 
preparation). It is used during aggressive 
encounters and is sometimes given by 
spectators. Perhaps it is used by a "frus- 
trated" aggressor — a bird intimidated but 
not cowed by his opponent(s) or by an 
aerial predator, or thwarted by poor visi- 
bility in the predawn twilight. Whatever 
the causation, the display is suited to func- 



tion as a long distance advertisement and 
specific identification of a territorial male. 

(2) Wing Rai.siug cum Buzz. A pattern 
which precedes or replaces Aerial Displays 
in some agonistic encounters is almost 
certainly homologous with Wing Raising 
of the larger congeneric species. Once a 
bird watched two others fight about 10 
meters away for over a minute, calling 
"tek" frequently. It began to jump up 
about 30 centimeters while lifting both 
wings, synchronously, fully over its back 
into the maximal position typical of the 
Aerial Display (revealing a large expanse 
of pale plumage). An abrupt, snapping, 
mechanical Buzz was produced as the 
wings either reached their full stretch or 
started down. After several such displays 
the bird paused, began Wing Raising cum 
Buzzes again without jumping, then did an 
Aerial Display with repeated stalls. 

In two other cases this Wing Raising 
cum Buzz was used by one participant 30 
or more seconds after a fight. In one case 
it was repeated twice by a perched in- 
dividual about 30 seconds before giving an 
Aerial Display, and in the other was given 
as an Aerial Dis]ilay when the bird flew 
up 2 meters, stalled in the usual fashion 
once but with the Buzz instead of calling. 
On yet another occasion. Wing Raising 
cum Buzz immediately preceded Aerial 
Display by an individual responding to the 
Aerial Display of a neighbor. In several 
dozen recorded Aerial Displays tcith vocal- 
izations, I have never detected a Buzz. 
Prol^ably the Buzz relates to the Wing 
Whirr of some larger species, but in M. 
mocuUrostris it has a ritualized relationship 
to Wing Raising. 

The wings are sometimes raised syn- 
chronously into the full upright position 
without a Buzz by a bird on the ground in 
an aggressive encounter, and this is per- 
haps the minimal form of the display. In 
the one case seen clearly, the bird gave a 
"tk" with each wing movement, then at- 
tacked its ojpponent. 



Evolution of Ground Tyrants • Smith and Vuillewnier 253 



(3) Otlier Wing And Toil Movements. 
Aggressive individuals facing an opponent 
often droop their wing tips nearly to the 
ground before and after fighting. The 
wings, and less often the tail, may be given 
little flicks. One aggressor cocked its tail 
up while letting its wing tips droop, with- 
out flicking either. Birds in these postures 
were silent and very likely to attack, dart- 
ing forward with rapid wing flicks, showing 
the upper wing surfaces to the opponent. 

One was seen to wing quiver slightly 
with each "tek" after landing from an 
Aerial Display. Within a minute it flew 
off — apparently to an agonistic encounter. 
Usually no wing movements accompanied 
"tek" calls. A slight and very rapid flick 
of the tail, however, often accompanied a 
"tek," especially if calling was rapid. An- 
other tail movement, a quick fanning parti- 
ally open and shut again, often preceded 
running. 

(4) Crown Ruffled. If I approached a 
fledgling and it called, the adult female 
might also call "tek" and ruffle her fore- 
head conspicuously. This is the second 
phase of crown ruffling in the larger spe- 
cies, most of which have an occipital spot 
of color which M. maculirostris lacks. 

Musc/sax/co/a brevicauda 

I studied this species {Muscigralla brevi- 
cauda of authors) at only one site in coastal 
Peru, from 23 to 26 January 1962. The 
population was locally dense, and there 
was much displaying. 

Vocal Displays 

( 1 ) A loud "tchek" ( Fig. 7a ) was often 
heard just as birds landed; it resembles in 
form and usage the Locomotoiy Hesitance 
Vocalization (LHV) of Sayornis plioebe 
(Fig. lie). The communicators were usu- 
ally watching as I walked through their 
territories, and probably most were males, 
since they also used the prolonged vocali- 
zations described below. In fact, the 
"tchek" appears to be incorporated into the 



prolonged vocalization, immediately pre- 
ceding the ultimate descending series. 

A slightly briefer fonn "tchk" (Fig, 7b), 
hard to distinguish by ear, occasionally 
intergraded with "tchek." "Tchk" was some- 
times weak, nearlv lacking its first element, 
with emphasis then falling on the third or, 
sometimes, the second element. It was also 
uttered on landing, or by perched individu- 
als making flight intention movements, and 
sometimes preceded landing or was re- 
peated two or three times during a very 
short flight. But usually it was in a short 
series as prologue to the prolonged vocali- 
zations given while perched. It may cor- 
respond to the Doubled Vocalization (DV) 
of S. plioebe. 

Occasionally in fights a harsh "zrrt" was 
used that may have been a version of 
"tchek"; it was not recorded. No forms 
similar to "tchek," "tchk," or "zrrt" are 
known from other Muscisaxicola species, 
except for the apparently rare LHVs. In 
usage, however, they at least partially re- 
place the two-peaked calls of M. albilora 
and other large species. The reduced visi- 
bility in the brushy habitat of M. brevi- 
cauda may require more complex calls for 
at least species identification. 

(2) There are at least three briefer 
vocalizations. One has an initial descending 
arm preceding a peak ( Fig. 7c ) that sounds 
like "tk" and is sometimes repeated two or 
three times in flight toward a perch, fol- 
lowed by "tchek" on landing. Occasionally, 
a similar series of "tk" followed by a "tchk" 
comes just after landing. In a short flight 
the series may be intermediate between 
"tk" and "tchk," with the third element of 
the "tchk" emphasized. 

Two other very brief calls sound like 
"tk" (Fig. 7d and e), but do not seem to 
form intermediates with it, although they 
do with each other. They are often inter- 
spersed among successive prolonged vocali- 
zations in flight, but do not form long 
series \\'ithout haxing inteipolated oc- 
casional "tk" calls of the first sort. The 
more obviously chevron-shaped element 



254 



Bulletin Museum of Comparatice Zoology, Vol. 141, No. 5 



ill 




a, 



,1 



tl 




—I i_ 



tlnit 



hi 



HW I 



Figure 7. Vocalizations and Pose of Muscisaxicola brevi- 
CGudo: 7a. "tchek" (compare with Fig. lid); 7b. "tchk" 
(this occasionally intergrades Into the previous vocalization); 
7c. "tk" (compare v/lth the elements of a "tchk"); 7d. "tk ' 
(compare with Fig. 6a]; 7e. "tk" (compare with Fig. 5a); 7f. 
prolonged series vocalization uttered while perched (all but 
the last unit of the introductory series has been omitted); 7g. 
prolonged series vocalization uttered during Aerial Display 
(all but the last unit of the introductory series has been 
omitted); 7h. customary pose while uttering a prolonged 
series vocalization from a perch. 



( Fig. Yd, compare with "tek" of M. inaculi- 
rostrls) is .sometimes uttered irregularly by 
a bird which alternates foraging with 
perching on a bush and uses this and no 
other calls; it also utters this call some- 
times on landing from a short flight. Oc- 
casionally a series of this version comes 
just after landing, followed by one or more 
"tchk" calls leading to the prolonged 
vocalization. 

Series of the second version (Fig. 7e) 
may be the more usual among successive 



prolonged vocalizations of a flying bird, 
and often continue as a bird completes a 
relatively long flight. Such long terminal 
series are broken by an occasional "tk" 
(Fig. 7c). On the whole, these last two 
versions of "tk" appear to correspond 
closely to "tek" and "tk" in the repertoire 
of M. maculirosiris, and must be related to 
Simple Vocalizations. 

(3) Prolonged series vocalizations (Fig. 
7f) are uttered either from perches atop 
cotton bushes or during Aerial Displays. 
Those in flight are usually, but not invari- 
ably, less complex ( Fig. 7g ) . All evidence 
indicates that the prolonged vocalizations 
correspond at least in part to what is 
usually called "song" in oscines, or to the 
Regularly Repeated Vocalizations of Soy- 
ornis and PyrocephaJus. They arc used in 
countercalling among conspicuous in- 
dividuals remaining within their apparent 
territories. 

Prolonged vocalizations may be stereo- 
typed elaborations of the less complex calls 
used by the same individuals during flight 
and landing. A "tchek" or "tchk" preceded 
by any of the "tk" vocalizations may cor- 
respond to a minimal form. The more com- 
plex flight form is a "tk" series plus a 
"tchek" with a harsh section inteipolated, 
and a set of five or more descending ele- 
ments appended (Fig. 7g). In the most 
complex (perched) version, the initial series 
is two to four "tchk" instead of "tk," plus 
yet another new feature, a section of 
ascending pitch matching the ultimate 
descending three elements. 

The whole vocalization sounds reminis- 
cent of a slurred version of the Aerial Dis- 
play call of M. mociilirostris, but the form 
of the prolonged vocalization is much more 
complex. If, however, the audible rise and 
fall of the second portion is equivalent to 
the "whee-oo" of M. macuUrostris, then 
the prolonged vocalization may be homol- 
ogous. Perhaps significantly, the "whee-oo" 
is at least rarely replaced by three chevron- 
shaped elements, and is in this form more 
similar. The greater complexity of the pro- 



Evolution of Ground Tyrants • Smith and Viiilleumier 255 



longed vocalization of M. brevicouda is 
consistent with the greater complexity of 
each of its other most common calls. 

Nonvocal Displays 

( 1 ) Wing Raising. No display seen was 
certainly comparable to Wing Raising of 
the other species. Once, however, after 
two individuals fought and chased, the 
pursuer perched and partly opened its right 
wing several times in rapid succession. 
Further, one incident before launching 
into Aerial Display also suggests that a 
display like Wing Raising exists, although 
I may not have seen its full form: A bird 
was repeating "tchek" fairly rapidly. It 
gradually extended its wings laterally, and 
without fully spreading the primary feath- 
ers, appeared to arch them slightly down- 
ward. It ruffled its back, fluffed its chest 
and crown, and uttered a minimal version 
of a prolonged vocalization (roughly: 
"tk-turrk"), then did a low display flight 
with one prolonged vocalization. 

(2) Aerial Display. A bird may launch 
into display silently or after a single pro- 
longed vocalization. Usually he climbs 
rapidly to about 6 or 7 meters above the 
vegetation, then flies with very full wing 
strokes. While flying he utters a prolonged 
vocalization, holding both wings stretched 
maximally upward during the final flourish 
of the call, lifting his head and body axis 
slightly upward but not into the vertical 
pose of other Muscisaxicola species. Titch- 
ing forward slightly, and usually veering at 
least slightly, he then repeats the perform- 
ance about three or four more times. Some- 
times he climbs continuously throughout, 
but if only one prolonged vocalization is 
uttered, the flight usually does not ascend. 
Except for the lack of full stalling (also 
omitted, but rarely, in the other species) 
and the deep wing stroke, the Aerial Dis- 
play is similar to that of other members of 
the genus, and to that of Pijwcephalus 
rubinus (which also does not usually 
stall). 



Other Display 

About 70 per cent of the prolonged 
vocalizations heard were from perched 
birds (the usual stance is shown in Fig. 
7h). Their bills open for each call and 
very widely for the ultimate flourish as 
the head is thrown back, and the tail and 
occasionally the wings are slightly flicked. 
While M. macidirostris never gives its full 
Aerial Display call from a perch, this usage 
by M. Irrevicauda may correspond to 
series of "tk" by an M. macidirostris before 
(or while apparently not quite ready to 
perform) an Aerial Display. 

Wing flicks are occasionally given by 
disturbed birds, and are very frequent if 
the individual is very excited. These are 
simple, small amplitude, forward and up- 
ward rotations of the wing tips. 

The tail of M. brevicauda is so short that 
adults look as if they should be fledglings. 
Nonetheless, it has a msty terminal bar 
which probably has some signal function 
such as making tail flicks more conspicuous. 
The only other movement seen to involve 
the tail is landing atop a bush, when the 
tail is cocked as if for steering or balance. 
The tail is probably at least partly fanned 
in the Aerial Display. 

Some plumage display has been men- 
tioned aboxe under Wing Raising. 

lib. Xo/m/s 

All X. ptjrope and X. irupero were seen 
in or immediately before the breeding 
season, and X. velata, X. cinerea, and X. 
striaticolJis only after the breeding season. 
Most published accounts report the birds 
as nearly or wholly silent, and undemon- 
strati\'e. I heard and saw little display. 

Xolmis pyrope 

Xobnis pyrope was observed in Chile 
near Isla Negra, Cerro Manquehue, and on 
Tierra del Fuego. 

Vocal Displays 

(1) A simple, chevron-shaped call, "pt" 
(Fig. 8a), sounds very like "tek" of Mtisci- 



256 Bulletin Museum of Comparative Zoology. Vol. 141, No. 5 



/ / / 



/ / 



/\ 



Figure 8. Vocalizations of Xo/mi's species: 8a. "pt" of X. 
pyrope; 8b. prolonged vocalization "wfieet whut T-T- 
wheeoo" of X. pyrope (an initial "pt" omitted); 8c. "puh" 
of X. irupero; 8d. several coupled vocalizations recorded in 
a fight of two X. irupero; 8e. vocalizations by one X. irupero 
after chasing another; 8f. "pew" of X. irupero; 8g. a nasal 
vocalization of X. irupero (compare with Fig. lid). 



saxicohi muculiro.stris and is probably an 
SV". I noted uses identical to those of that 
species (q.v. ), except that X. pyrope was 
not seen to mob a predator. This "pt" may 
beWetmore's (1926: 301) faint 'V/c/c iic/c." 

(2) A very brief and relatively high 
pitched "tseet" was heard several times in 
a face-to-face aerial fight. Another time, 
an X. pyrope pursued a conspecific bird 
closely through some bushes, then immedi- 
ately after perching called: "tseet pwut, 
pwut pwut, put." These "pwut" and "put" 
w ere probably related to the "pt." "Tseet" 
may correspond to at least the "teek" and 
"tsk" of M. alhilora; if so, all are SV forms. 

(3) An encounter between an apparent 
pair and a third individual was seen on 
the boundary of a transient territory during 



spring migration. The third indixidual and 
one other came to a stand-off, both uttering 
the l)rief "pt" frequently and countercalling 
with bouts of a much more complex, pat- 
terned vocalization: "pt wheet whut T-T- 
wheeooo" (Fig. 8b, which omits the initial 
"pt"). The ultimate portion is closely 
similar to the prolonged \'ocalization of M. 
macuUrostris, and the whole follows the 
basic pattern of a series of brief calls and a 
flourish. 

(4) A weak, low, monosyllabic whistle 
is said to be characteristic (Goodall et al., 
1957: 151). It may correspond to the 
"tseet" (IPV) of M. alhilora, but I have 
not heard it. 

Nonvocal Displays 

None was certainly identified. In the 
single stand-off encounter, both partici- 
pants sometimes raised both wings quickly 
and slightly above the back, probably a 
display similar to the Wing Raising of 
Muscisaxicola species. The individual 
which called after terminating a non- 
aggressive chase had a conspicuously iiif- 
fled throat. During that chase the white 
in the tail of both individuals was especially 
conspicuous. 

Other Xolmis Species and Neoxo/m;s 

(a) In northwestern Argentina I found 
contiguous territories of Xolmis irupero in 
dry, gravelly river beds below the foothills 
of Salta and Jujuy, and saw several terri- 
torial chases. When perched, each partic- 
ipant would utter a "puh" (Fig. 8c) at 
long intervals; this call is rather prolonged 
and flat, with a terminal increase in ampli- 
tude. The usage suggests the briefer "pt" 
of X. pyrope, but the length suggests the 
"sect" of M. alhilora. Sayornis saya uses a 
prolonged and rather flat IP\' in many ago- 
nistic situations, and it and S. nigricans use 
IPX's in situations in which S. phoehe usu- 
ally uses an SV. 

Immediately after chasing an intruder, 
the victor usually gave variable calls, often 
suggesting the initial portion of the pro- 



Evolution of Ground Tyrants • Smitli and Viiilleiimier 257 



longed vocalization of X. pyrope (see Fig. 
8e). One such victor landed and gave a 
prolonged "pweeeeet" plus three couplets 
of modified "puh" calls resembling the calls 
of an X. pyrope after a chase (see above). 
Some calls during fights had coupled 
elements (Fig. 8d) and one was a broad 
chevron ("pew," Fig. 8f). These all ap- 
pear more like SV than the simple "puh," 
and most differ primarily in duration and 
slope. Calls recorded from one fight are 
different, a set of eight very brief, nasal 
elements (Fig. 8g) quite reminiscent of an 
LHV. 

(b) Little is known about the displays 
of other Xolmis species. Hudson (1920: 
142) remarked that X. coronoto has a "long, 
low, whistle." Wetmore (1926) has heard 
a "faint swee" from X. cinerea, as well as 
a 'little whistled song," used just at dawn 
and probably corresponding to the pro- 
longed vocalization of X. pyrope. Of X. 
riibetra, he says that males "at intervals 
flew up to make a metallic rattle with their 
wings as they turned abruptly and dropped 
to the ground." This is the only indication 
of which I am aware of some sort of flight 
display and Wing Whirr or Buzz in the 
genus. It may be similar to flight displays 
in Knipolegus and related genera (Smith, 
in preparation) of a lineage probably fairly 
closely related to die genera discussed 
herein. 

(c) I have seen X. fumigata only in the 
post-breeding season, and the birds did not 
display. Vuilleumier (personal communi- 
cation) has heard a loud "tew" from forag- 
ing individuals of X. strioticoUis in July, 
in Ecuador. In Argentina he found one 
individual in April 1965; it uttered a loud 
"they-tew-thew." Other accounts of the 
call of this species (e.g., Todd and Car- 
riker, 1922; Koepcke, 1964) resemble his. 

One individual of an apparently un- 
described population of this species-group 
was observed by my wife and me on 15 
June 1966, at about 4000 meters on the 
northwest slope of \'olcan Cotopaxi in 
Ecuador. The bird was medium brown 



above, paler buffy below, pale yellowish 
on the throat, and had a very faint super- 
cilium. Extremely faint traces of broken 
streaking marked the lower throat and 
flanks. The bill was black. The plumage 
was ver)^ fluffed, hiding most of the wings 
so that it was impossible to tell if these 
were completely unmarked, but they could 
not have had prominent bars. In shape, 
size, and color the bird was very close to 
specimens I have examined of X. fuscorufa 
of southern Peru and Bolivia, but lacked 
prominent wing bars. According to Hell- 
mayr (1927), the local X. pernix of the 
Santa Marta range in Colombia is the 
population phylogenetically closest to X. 
fuscorufa, but it has much less prominent 
wing bars and a more reddish coloration (I 
have seen only the one skin of this popu- 
lation in the M.C.Z. ). A likely explanation 
of our bird is that it was a member of a 
deme intermediate between these two dis- 
junct populations, and that they are ex- 
treme members of a linear series of 
populations. (The genus has another sup- 
posed rare and local population, X. sig- 
nata, in addition to X. pernix and our 
bird. As it is an "island" genus, such 
populations are perhaps to be expected.) 
I shall refer to this bird as X. (fuscorufa/ 
pernix? ) . 

It perched silently atop a tall bush for 
several minutes, then repeated a loud 
"cleeoo" (Fig. 10a), building quickly to 
a rate of 29 per minute and maintaining 
this for over five minutes, like the BRV of 
a Sayornis species. There was slight vari- 
ation in at least the slope of the terminal 
arm of the call, which is markedly similar 
to the RR2 of RRV ("song") of S. nigri- 
cans (particularly an individual recorded 
in Panama, Smith, in press, 1 ) . After seven 
to eight minutes, calling stopped, and the 
bird dropped to the grass where it stood 
like a Turdus thiTish for some tens of 
seconds before dashing off. We were not 
able to find it again. 

(d) Neoxolmis rufiventris, closely re- 
lated to the species of Xolmis, is said to 



258 Bulletin Museum of Com pa rat ice Zoology, Vol. 141, No. 5 



^ 



^ If 



_i 1 1 i_ 




_l I I I L. 




Figure 9. Vocalizations and Pose of Agriornis species. 9a. "pyuk" of A. montana, Argentina; 9b. "pyuk" of same indi- 
vidual A. montana; 9c. "wheet hyou" of gray Agn'ornis, Ecuador, (also A. montana?); 9d. "t-eek" of A. Itvida; 9e. 
"t-eek-ek" of A. livida; 9f. pose of A. livida while uttering "t-eek" calls. 



have a "long, low, plaintive whistle" ( Hnd- 
son, 1920: 138). 

lie. Agriornis 

I have seen two of the five species, but 
never in dense populations. Both have a 
loud whistle which probably serves some 
"song" functions, since birds utter it from 
atop shrubs or other relatively tall vegeta- 
tion at long, irregular intervals (often 
about 45 to 80 seconds). 

Agriornis montana, recorded in Tucu- 
man, Argentina, has the briefest vocal- 
ization, a relatively low pitched "pyuk" of 
SV form (Fig. 9a\and b). Goodall et al. 
( 1957 ) say this is used in the early morn- 
ing, which is when I recorded it, and con- 
sider the species more vocal than A. livida 
in Chile. Dark gray birds on the northeast 
slopes of Mt. Pichincha in Ecuador had 
the longest call (Fig. 9c), a clear "wheet 
hyou" with a form like a Sayornis saya 



RRl song unit (see below, and Fig. llh). 
A bird using it also foraged while under 
observation, and sometimes countercalled 
with another individual using the same 
vocalization. Vuilleumicr has heard a 
nearly identical call from A. montana in 
the Potosi Department of Bolivia (see his 
discussion in Chapter 1). 

The corresponding vocalization of A. 
livida in central Chile is somewhat like a 
Sayornis phoehe IPV: "t-eek" or "t-eek-ek" 
(Fig. 9d and e, with pose of bird between 
calls in Fig. 9f). Fully grown, nearly in- 
dependent fledglings were found at Los 
Molles, Chile, from 7 to 9 Januaiy 1962. 
They sometimes perched and called a flat- 
ter version lacking a terminal peak: "pwut." 
It sometimes \\'avered or was harsh. One 
perched alone once and uttered a loose 
series of weaker "peet" calls that became 
slightly harsher ("pwt") before it flew off. 
Later it landed near me, looked at me for 



Evolution of Ground Tyrants • Smith and Vuilleumier 259 



20 seconds, then with a faint, harsh "pwt" 
began to forage. Still later, when an adnlt 
came the fledgling gaped toward it and 
repeated "peet" two to three times per 
second, the "peet" getting harsher until 
the adult flew away. 

The known vocalizations of adults of 
these Agriornis populations are not closely 
similar to one another. All, however, re- 
semble different vocalizations of Musci- 
saxicola and Sayornis species which are 
used (at least in part) similarly. The one 
visible display known is an Aerial Display 
by an A. oUricouda seen by Vuilleumier 
(personal communication) in La Paz De- 
partment, Bolivia. The bird circled silently, 
alternately rising to a partial stall and 
dropping forward on closed wings, as in 
an incomplete Aerial Display of a Mtisci- 
saxicoJa species. 

Wetmore (1926) heard "sharp squeaky 
notes" (perhaps BVs) from Agriornis 
murina during a pursuit. 

lid. Ochthoeca 

Ochthoeca fumicolor 

I observed this species in Ecuador on 
Cerro Atacaso in 1959 and 1966, and on 
Volcan Cotopaxi in 1966. The following 
vocalizations have been recorded. 

(1) A very variable, loud, clear, abrupt 
whistle: "kleeip," "kleeep," or "kleeeh" 
(Fig. 10c, d, e, and f). The variations 
have not been correlated with different 
usages, but are comparable to variations of 
the very similar IPV of Soyornis phoehe. 
The call is uttered at long intervals by some 
foraging birds while perched, or sometimes 
in flight. Sometimes two countercall with 
it: e.g., immediately after an agonistic 
encounter, both participants uttered it be- 
fore one flew away. An adult repeated it 
nearly every time it fed its fully grown 
fledglings, and spanned all of the recorded 
variations. Most calls of the fledglings 
(e.g.. Fig. lOg) were similar to one sort 
of the adult's, and much fainter. All usages 
strongly suggest the IPV of Sayornis saya. 



( 2 ) A less loud "tee-oo" ( Fig. lOb, rarely 
just the descending arm is uttered) is 
nearly indistinguishable in sound from the 
one known call of Xolmis {fuscorufa/per- 
nix?). At least once, two individuals sepa- 
rated after one agonistic encounter, then 
countercalled with this before the next. 
It is very similar in sound and form to the 
terminal calls of Aerial Displays in Miisci- 
saxicola and the prolonged call of Xolmis 
pyrope, as well as somewhat similar to the 
one known call of the Agriornis species 
recorded in Ecuador. All apparently relate 
to RRV units of Sayornis. 

(3) Ochthoeca fumicolor also uses pat- 
terned series of one to three variable 
"kleep" vocalizations followed by about 
eight to ten simpler "klee" calls (Fig. lOh, 
probably also IPV or RRV forms). The 
latter are nearly invariable in form, and 
nearly evenly spaced about 0.4 seconds 
apart, with a slightly longer pause before 
the ultimate one. In series that had as few 
as two "klee" units these sounded more 
like "tee-oo." In 1966 the only series we 
recorded were during and after an ago- 
nistic encounter, as two participants 
countercalled at irregular intei'vals. They 
eventually shifted from using both "kleeip" 
and the patterned series to just "kleeip." 
One individual in 1959 uttered short series 
while foraging near a calling O. rufipecto- 
ralis (sec below). 

(4) Two other vocalizations, recorded 
during fights, both resemble calls used in 
fights by at least M. albilora. One (Fig. 
lOi) sounds like "pw-pwt" or "pw-pw-pwt," 
the other ( Fig. lOj ) like "twee tik." Both 
may be uttered in bursts, and at least the 
former is sometimes used in flight with a 
Wing Whirr. Both may be LHV forms, 
and have similarities with the LHV of S. 
phoehe ("twh-t," see Fig. lie). 

Some of O. fumicolor s nonvocal displays 
arc also known. During and after agonistic 
encounters there was much wing and tail 
flicking that was rapid during series calls, 
but there was some also with at least 
"tee-oo" calls. When foraging with or 



260 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 





J 



^. 



u 



1 ^^^ 



.\ 





Figure 10. Vocalizations of Xo/m/s (fuscorufo/pernix?) and 
Ochthoeco species: 10a. "cleeoo" of Xo/mis (fuscorufa/ 
pernix?); 10b. "cleeoo" of O. fum/'co/or; 10c. through 1 Of. 
variations of the abrupt whistle of O. lumicolor-. "kleeip," 
"kleep," "kleeeh," and "kleeeh," respectively; lOg. a 
similar vocalization of a fledgling O. fum/'co/or; lOh. "klee ' 
vocalization of O. fum/co/or; lOi. a vocalization recorded 
from a fight of two O. fum/co/or; lOj. another vocalization 
recorded from a fight of O. fum/'co/or (compare with 
Fig. lid); 10k. "tyeh" of O. /eucophrys (Peru); 101. "tyee" 
of O. leucophrys (Argentina); 10m. "tee tee ti," recorded 
from a fight of two O. /eucophrys (Argentina). 



without occasional "kleep" type calls, there 
were occasional tail flicks (always up/ 
down), especially on landing. A Wing 
Whirr in several flights during at least one 
agonistic encounter had tw^o forms: several 
single or paired snaps about 0.2 seconds 
apart, or a quick burst of about seven 
similar sounds in about 0.4 seconds 
(analyzed from recordings). The individ- 
ual foraging near an O. rufipectoralis gave 
very noticeable bursts of whirring every 
time it flew. 



Other Ochthoeca Species 

A single foraging O. rufipectoralis was 
observed for an hour on 7 August 1959, on 
Cerro Atacaso. It occasionally uttered a 
rather faint "cleeoo" (not recorded) which 
I could distinguish from the "tec-oo" being 
used by an O. fiimicolor primarily by 
amplitude. That these two species and the 
Xolmis {fuscoruf a / pernix?) , all in the same 
general region, should have such remark- 
ably similar calls is probably due to more 
than close phylogenetic relationship. The 
species probably have some organized 
social relationship (see Moynihan, 1968, 
for various possible interspecies relation- 
ships). Perhaps it is significant that while 
the only two members of Ochthoeca in the 
copse during that hour were of different 
species, both kept displaying but had no 
encounter. 

One other vocalization was heard when 
the O. rufipectoralis was especially active: 
a long, rapid, series of brief "pt" calls, 
terminating in two "cleeoo" calls. If this 
corresponds to the series call of O. fiimi- 
color the reversed sequence in the two 
species may be significant in the light of 
the unusual interspecies similiarity of the 
aforementioned vocalizations. The brief 
"pt" calls seem likely to be SVs, and may 
replace the IPV in this species — such a 
substitution occurs among the species of 
Sayornis. 

Finally, the O. rufipectoralis tail flicked 
less than its slightly larger congeneric as- 
sociate, and Wing Whirred only twice, 
faintlv. However, all its flights undulated 
like Wing Whirr flights, being broken into 
segments by recurrent pumping of the 
relatively long tail. 

I recorded O. leucophrys near Lima, 
Peru (September 1962), and in northwest- 
em Argentina ( November 1962 ) . The lone 
individual in Peru foraged with an oc- 
casional abrupt, SV-like "tyeh" (Fig. 10k) 
when perched or during flights of 6 to 25 
meters between bushes. On calling while 
perched, it would give a slight wing flick 



Evolution of Ground Tyrants • Smith and Vuilleumier 261 



and single tail flick. Once it called "tyeh 
tuh tuh tuh tuh" in flight as it approached 
a perch. 

At the Argentine site two pairs of O. 
Jeucophrys were in a narro\\' quebrada, and 
one individnal kept intruding into the area 
of the other pair. Several times one chased, 
fought, and expelled the intruder, and in 
each encounter one used brief bursts of: 
"tee tee ti" (Fig. 10m). These resemble, 
respectively, an SV, BV, and SV of Sayor- 
nis phoebe, and, to a lesser extent, calls 
recorded from flying Xolmis and Musci- 
saxicola. Between intrusions the defender 
foraged with an occasional "tyee" (Fig. 
101), slightly more prolonged and more 
IPV-like than the "tyeh" of the Peruvian 
bird. Vuilleumier watched two individuals 
foraging near Mitotambo, Peru, on 27 May 
1965, and one or both uttered "wheet" and 
"wheeyet," calls which may correspond to 
the "tyeh" (SV) and "tyeee" (IPV) I have 
recorded. (It is very difficult to correlate 
onomatopoetic descriptions of different ob- 
servers, since bird vocalizations do not 
correspond closely to the vowels and con- 
sonants we must use to describe them; 
Vuilleumier and I each heard calls in 
similar circumstances which fell into two 
categories by length and were generally 
similar in pitch.) 

Vuilleumier ( personal communication ) 
heard an O. cinnamomeiventris call oc- 
casionally while foraging below the Carpish 
Pass, Peru, on 28-30 May 1965: "long, high- 
pitched seee, very reminiscent of the call 
of Colorhamphus parvirostris." 

COMPARATIVE SUMMARY 
AND DISCUSSION 

The known behavioral similarities of the 
above genera are here summarized, and are 
compared with the behavior of the genera 
Sayornis (for which all references below 
are to Smith, 1969, and in press 1 and 2) 
and Fyroceplialus (for which references 
are to Smith, 1967, plus subsequent work 
in preparation). 



Habitat preferences vary along a con- 
tinuum transcending generic limits. While 
some Ochthoeca species inhabit dense 
forests, others live on the wooded or bushy 
fringes of the open, grassy paramo, and 
devote much of their foraging attention to 
it. In opcness and general distribution of 
bushes for perching, the latter habitats are 
comparable to those of Sayornis and Pyro- 
cepholus. Among Xolmis species, even 
members of the fiimigata species-group are 
not found in dense forest, and range pri- 
marily from edge to semi-open habitats 
with tall perches. Other Xolmis species 
carry this trend further, and at least two 
are as terrestrial as some Muscisaxicola 
species, although perhaps not identical in 
their modes of foraging. Extremes of open 
habitat preference are reached in both 
Agriornis and Muscisaxicola, but in all 
cases some feature provides elevated look- 
out posts. The less extreme members of 
even these genera have breeding habitats 
with abundant bushes or some trees. 

Attention in searching tends to be di- 
rected downward in some Ochthoeca spe- 
cies, in Sayornis saya, and at least season- 
ally in other Sayornis species. This feature 
is particularly obvious in Xolmis, most 
species of which perch relatively high for 
terrestrial foragers. The larger, ground- 
dwelling Muscisaxicola species direct most 
of their searching attention downward, al- 
though even they have some tendency to 
scan the air for flying prey. Tliroughout 
the group as a whole, locomotoiy patterns 
show obvious adaptations to the form of 
searching, from a tendency to fly close to 
the ground to the development of rapid 
hopping and running. 

All of the species do some or much 
foraging by gleaning, and possibly a tend- 
ency to glean is necessary in the e\ olution 
of terrestrial habits, since most prey in 
open countiy is likely to be on the ground 
or clinging to plants. Particularly where 
open habitats are ^^'indy, flying inverte- 
brates are relatively uncommon. Yet aerial 
flycatching persists. It may be of major 



262 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



significance only in Soyornis, Pyrocephahis, 
and some Oclithoeca species, but all of the 
species under consideration do it. It is 
behavior that characterizes the tyrannids 
among the other terrestrial passerine in- 
sectivores \\'hich share their habitats. An- 
other such feature is the tendency of 
Muscisaxicola and some Xolmis species to 
alternate short dashes and/or flights with 
pauses to scan for prey; in central Chile 
and coastal Peru, Anthus species (motacil- 
lid), and Geositta and other terrestrial 
fumariids sympatric with one or more 
Muscisaxicola species usually walk con- 
tinuously while foraging. The Old World 
flycatching muscicapids ( genus Oenanthe ) , 
^vith which Muscisaxicola species are con- 
vergent, use the same foraging tactics, 
while their sympatric passerine insectivores 
(at least in Iceland, where these are 
Anthus and MotaciUa [personal observa- 
tion] ), are also "walkers." It is as if a fly- 
catcher, of whatever phylogenetic affili- 
ation, in taking to terrestrial foraging, 
preempts this tactic. 

All members of the assemblage build 
open cup nests. They employ available 
opportunities for concealment, placing their 
nests in vegetation (grass clumps or 
bushes), in crevices or burrows, or on 
partly concealed ledges. Sayornis species 
use only the last, including artificial sub- 
stitutes like bridges and buildings, and 
Oclithoeca nifipectoralis has been known 
to use a similar site. Muscisaxicola is the 
only genus restricted to terrestrial nesting 
(if we consider ledges as being off the 
ground), and Agriornis montana and 
Neoxolmis rufiventris may be the only other 
species nesting on the ground (some popu- 
lations of Xolmis cinerea may nest in holes 
in banks). That closely related species 
should differ in nesting on the ground or in 
bushes suggests that evolution of ground 
nesting has not been difficult for open 
country flycatchers. 

The nondisplay behavior patterns do not 
appear to set any of the genera apart, but 
show a general trend from some adaptation 



to open environments in Sayornis, Pyroccph- 
aliis, some Xolmis, and some Ochthoeca 
species, to considerable adaptation in 
Muscisaxicola, other Xolmis species, and 
Agriornis. By "trend" I do not mean to 
imply that any of these genera is a part of 
the phylogenetic lineage of another. Ob- 
\'iously, all are contcmporaiy end-products 
of adaptation to different habitats, but they 
serve to show the nature of possible stages 
in the evolution of terrestrial forms from 
less terrestrial ones. 

Comparison of the displays is less easy 
than of the nondisplay behavior. It can 
best be done by comparison with the dis- 
plays of the related genus Sayornis, in 
which all displays of at least S. phoehe 
appear to be known, and the relationships 
among thos(> displays are relatively well 
understood. No attempt is made here to 
repeat detailed descriptions of the form 
and usage of Sayornis displays; they are 
cited below only insofar as is necessary to 
establish points of comparison or contrast 
with the other species. Oversimplification 
is inevitable, and most of the following 
general statements about form and usage 
of displays can be qualified by a variety 
of exceptions. The exceptions do not ap- 
pear to contradict the conclusions, but a 
reader wishing a more complete compari- 
son must consult the more detailed papers 
on S. phoehe and the genus Sayornis. 

One important problem is that the 
amount of use of different vocalizations 
almost certainly varies with the stage of 
the breeding cycle sampled during the field 
work, as has been shown for S. phoehe. 
Since all the specific display repertoires 
appear comparable in many respects, sea- 
sonal shifts can likely be predicted from 
what is known about S. phoehe. 

(1) Simple Vocalization (SV). In S. 
phoehe this sounds like "tp" (Fig. 11a). 
It is used in a varietv of circumstances, 
most of ^^'hich are agonistic or potentially 
agonistic: foraging while patrolling, re- 
sponses to predators, and agonistic or 



Evolution of Ground Tyrants • Smith and Vuillcumier 263 



potentially agonistic interactions of mates. 
The SV of S. nigricans and the "peent" 
vocalization of Pyrocepluilus ruhinus are 
full chevrons, similar in most known 
usages. 

In Muscisoxicolo, at least some larger 
species divide the SV into a number of 
intergrading forms, each with its own range 
of usage. Thus, in M. olhilora, "teek" (Fig. 
Ic) is usually used in approach toward 
an interaction which may be agonistic, 
whereas "tchk" and "tk" (Fig. 2d) are 
used in grappling fights, and almost all 
forms are heard when mates meet during 
or immediately after an agonistic situation. 
M. frontalis and M. flavinucha appear to 
have similar SVs, M. rufivertex and M. 
macloviana may, while M. capistrata either 
lacks such calls or uses them primarily at 
other parts of the breeding cycle than 
were observed. 

Mtiscisaxicola maculirostris has a chev- 
ron-shaped homologuc ("tek," Fig. 6a) 
which it uses much more abundantly than 
the SVs of any of its larger congeners. The 
indistinct variations of "tek" appear to cor- 
respond to the different SV forms of M. 
aJhiJora. The yet smaller M. brevicauda 
has only comparable "tk" forms (particu- 
larly Fig. 7d), and restricts their usage (see 
below, under LHV). 

All Miiscisaxicola species employ forms 
closely related to SVs (and here grouped 
\^'ith them) in short series or bursts in 
circumstances suggesting Chatter Vocal- 
ization (CV) usage of Sayornis species. 
These usages are either by a bird associat- 
ing with its mate without attempting other 
activities, or by males before or in Aerial 
Displays. The one obser\'ation of CV-type 
calls by a lone M. frontalis apparently pros- 
pecting for a nest site is significant, since 
it is typical of CV usage in many better 
known tyrannid genera, and in the three 
Sayornis species and P. rid)inus. The forms 
resemble the chevron series C\^ of S. nigri- 
cans (Fig. lid), and similar variants of 
the CV of S. say a. 

Xolmis pyrope has a "pt" (e.g.. Fig. Sa) 



U 



/l/l /I l\ 



J I I L. 



C d 

I I I I I 



-J I I 




f g 

-I 1 1 I 1 I I I I I I I I 



/> 



\/v- 



/Vs^h 



- h 



-J I I I I L 



J 1 1 I 



Figure 11. Vocalizations of Sayornis species: 11a. Simple 
Vocalization (SV) of S. phoebe; lib. Initially Peaked Vocali- 
zation (IPV, one variant) of S. phoebe; lie. Bipeaked Vo- 
calization (BV) of S. phoebe; lid. A portion of a CInatter 
Vocalization (CV) of S. nigricans: lie. Locomotor/ Hesitance 
Vocalization (LHV) of S. phoebe; 1 1 f. A unit of Cfiatter Vo- 
calization (CV) from a series of S. phoebe; llg. An Initially 
Peaked Vocalization (IPV) of S. soyo; llh. An RRl of the 
Regularly Repeated Vocalization (RRV) of S. soyo; lli. An 
RRl of the RRV of S. nigricans (Arizona); llj. An RR2 of the 
RRV of S. nigricans (Panama) (compare with Fig. 10a). 



which corresponds to the SV of M. macu- 
lirostris, and X. irupero has similar but 
more prolonged calls (e.g., "pew," Fig. 8f) 
as well as coupled elements which are SV 
or CV. Some calls of other Xolmis species 
described in the literature appear to be 
SVs. 

In Agriornis, at least the "pyuk" of A. 
montana is of S\^ form and usage. Calls 
recorded from Ochthoeca leucophrys dur- 
ing fights ("tee" and variants, Fig. lOm) 



264 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



and apparent patrolling ("tyck," Fig. 10k) 
are probahK* S\'s. 

(2) Initially Peaked Vocalization (IPV). 
In Sayomis phoehe IPV (Fig. lib) is very 
rare and variable. It is sometimes used by 
males before going to roost, after an 
agonistic encounter, or when thwarted in 
an agonistic situation. S. nif:,ricans uses its 
honiologue more than S. phoehe, although 
in similar circumstances as well as in some 
patrolling and with RRV, and in S. saya 
the IPV (Fig. llg) virtually replaces the 
SV. Thus, \\'ithin Sayomis one species uses 
primarily S\' and little IPV, another uses 
both, and the third primarily IPV. Similar 
patterns may recur in the other genera. 

Muscisaxicola alhilora has a slightly bi- 
peaked "sect" (Fig. le and f) which occurs 
between attacks in agonistic encounters 
like an IPV; its "tseet" (Fig. la) is also 
IPV-like. M. frontalis has two correspond- 
ing calls ( Fig. 4a and b ) . Both species also 
have SVs. M. capistrata may have only 
IPV forms (Fig. 5a, b, and e). A "tee" 
heard from M. rufivertex with Wing Rais- 
ing may be either IPV or SV. But if M. 
maculirostris has an IPV (perhaps the 
squeak. Fig. 6b, by an individual on 
ceasing to approach a fight), it is very 
rarely used, whereas SV is used abundantly. 

In XoJmis irupero, "puh" (Fig. 8c) is 
probably an IP\^ as is the "t-eek" ( Fig. 9d 
and e) of patrolling Agriornis livida. The 
most Sfl|/o;-?i/.s-like IPVs are found in 
Ochthoeca, the genus most like Sayomis 
in habits and plumage. The variable 
"kleeip" (Fig. 10c to lOf) used in patrol- 
ling, countercalling, and after agonistic 
encounters is the IPV of O. fumicolor, a 
species with no known SV. Some other 
calls in this genus may be either IPV or 
RRV forms (see below). 

(3) Calls of Nestlings and Fledglings. In 
Sayomis, nestlings of S. phoehe beg with 
an SV-like "tee," those of S. saya primarily 
with IPV forms, and those of S. nigricans 
with both, thus roughly corr(>s]:)onding to 
differences in the frequency of usage bv 
adults of SV and IPV. The begging of 



young Muscisaxicola maculirostris and M. 
capistrata suggests a similar correlation in 
that genus, and in Ochthoeca fumicolor 
begging fledglings used only IPVs — as did 
adults. It does appear as if either IPV or 
SV can serve nearly the same functions in 
different species, and that which is more 
prominent is not a generic characteristic. 

(4) Regularly Repeated Vocalization 
(RRV). Each Sayomis species has a char- 
acteristic "song" of two units (called RRl 
and RR2) repeated in patterns for up to 
thirty minutes at a time. In S. saya, IPV 
units make up the bulk of most "song" 
bouts. The most closely similar perform- 
ance heard from any of the species de- 
scribed here was the bout in which the 
single Xolmis (fuscorufa/pemix?) repeated 
regularly a "clccoo ' ( Fig. 10a ) with a 
form like the RR2 of S. nigricans (Fig. 
llj). Ochthoeca fumicolor has a closely 
similar call ("cleeoo," Fig. 10b), recorded 
in countercalling l^outs, and a regularly 
repeated "klee" (Fig. lOh) which is either 
an IPV variant or a second RRV unit (see 
Fig. Hi). O. rufipectoralis has a "cleeoo," 
and the "tyee" (Fig. 101) of O. leucophrys 
is similar to an RRV unit (Fig. Hi). It is 
likely that the "wheet hvou" ( Fig. 9c, com- 
pare with Fig. llh) of gray Agriomis in 
Ecuador is an RRV unit, but it was never 
heard in series. 

The calls referred to as "prolonged vocal- 
izations" in several species of Muscisaxi- 
cola and Xolmis pyrope are like the RRV 
of Pyrocephalus ruhinus. Although they 
are not regularly repeated for long, even in 
Aerial Displays, they are used in counter- 
calling. Most terminate with "cleeoo" or 
"clee-ip," and those recorded (Figs. 6g, 6h, 
6i, and 8b) appear to be simplified \'ersions 
of the O. fumicolor "cleeoo" (Fig. 10b), 
preceded by a series of brief vocalizations 
which are usually SV- or CV-like. The 
complex prolonged vocalization of M. 
hrevicauda (Fig. 7f and g) appears to be 
a secondary specialization of the common 
form, conxergent in harshness with the 
RRl of S. phoehe. 



Evolution of Ground Tyrants • Smith and Vuilleumier 265 



Regular repetition, particularly in pre- 
dawn bouts, as a characteristic RRV usage, 
seems lost in the terrestrial genera, and its 
use in an Aerial Display is correspondingly 
increased. Sayornis species sometimes, and 
P. nihinns frequently, use their RRVs in 
flight displays, and also have predawn 
calling patterns, and occasional daytime 
bouts. 

(5) Bipeaked Vocalization (BV). In 
Sayornis phoehc and S. nip^ricans this has 
the form of a rather abrupt and multi- 
peaked IPV (Fig. lie) and sounds like 
"T-keet." It is uttered by the aggressor in 
agonistic encounters and occurs in some 
Flight Displays. 

In Miiscisaxicola the most similar vocal- 
ization is also IPV-like, and forms inter- 
mediates with IPV. The only larger species 
it has been recorded from is M. albilora 
("tseek," Fig. lb), used by birds flying 
toward their mates, and in agonistic en- 
counters between pairs. In M. maculirostris 
the closest approximation was a "tek" 
variant (Fig. 6c) used by a bird approach- 
ing a fight with considerable hesitance. 
Similarly, in M. brevicauda a "tk" variant 
(Fig. 7e) may approximate this call. 

In Agriornis, only the call described by 
Wetmore from a pursuit in A. miirina may 
be a BV. A "tee" (Fig. 10m) used in a 
territorial fight by Ochthoeca leucoplirys 
appears BV-like. 

(6) Locomotory Hesitance Vocalization 
(LHV). Sayornis phoehe has a call 
("twh-t," Fig. lie) used primarily on land- 
ing when there is a conflict of tendencies 
to continue flying or to take a perch. In 
early spring it is used during patrolling, 
but thereafter most commonly by males 
approaching their usually aggressive mates, 
or by males being attacked by their mates. 
Less commonly, a bird appearing "nervous" 
in any potentially agonistic situation may 
sometimes utter it on landing from a short 
flight. S. saya has a similar form known 
only from male-female disputes. A very 
similar "twee-tk" ( Fig. lOj ) heard in fights 
of Ochthoeca fumicolor is the closest ap- 



parent homologue in the other genera. A 
nasal call from a fight between two Xolmis 
irupero (Fig. 8g) may be an LHV. 

In Muscisaxicola alhiJora, apparent LHV 
forms were recorded only in a few greet- 
ings between mates. It is not certain that 
M. maculirostris has a homologue, but in 
my brief field work with M. brevicauda, 
an LHV ("tchek," Fig. 7a) was heard very 
commonly. It was used by patrolling 
males, almost always on landing from a 
flight (as in S. phoebe). No greetings were 
seen. Within their respective genera, both 
M. brevicauda and S. phoebe are promi- 
nent for having harsh calls. Even the CV of 
S. phoebe is harsh, and intergrades with 
its LHV. In the other species, CV may 
replace the LHV in some usages, and the 
SV and/or IPV in others. 

(7) Doubled Vocalization (DV). A rare 
display of S. phoebe, this is usually a 
couplet of brief chevron-shaped calls. It 
may represent a partial remnant of the 
CVs of the other Sayornis species, and is 
used in some instances in which an LHV 
might seem appropriate. Only M. brevi- 
cauda appears to have a call ("tchk," Fig. 
7b) which corresponds in usage, and this 
species is like S. phoebe in having a fre- 
quently used LHV. 

(S) Flight Display. The Flight Display 
of Pyrocephalus is similar both in its form 
and its usual vocalization to the Aerial Dis- 
plays of Muscisaxicola species. The Aerial 
Displays of M. brevicauda differ primarily 
in having a very full wing stroke, but De- 
Benedictis (discussed in Smith, 1967) has 
seen full wing strokes used by some in- 
dividuals of Pyrocephalus. The conspicuous 
Flight Display of S. phoebe is probably 
comparable, but may usually be relatively 
fast and erratic; few detailed obsen'ations 
exist. All Sayornis species have fluttering 
flights with CVs and RRVs and/or IPVs, 
which differ primarily in lacking stalls and 
in usually being oriented toward potential 
nest sites. Aerial Displays may also be 
found in at least Xolmis and Agriornis. 

(9) Wing Raising. The elaborate Wing 



266 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



Raising displays known in agonistic en- 
counters of Muscisaxicola species inter- 
grade with the Aerial Display. X. pyrope 
may also have a Wing Raising display, but 
there is no evidence for homologous dis- 
plays in the other genera. S. phoehe ap- 
parently lacks such a display, although its 
Wing Fluttering display may assume some 
of the functions of Wing Raising. Males 
of both S. saya and S. nip-icans have been 
seem to \\^ing Raise briefly at the end of 
short fluttering display flights, in what ap- 
peared to be pair-bonding contexts. 

(10) Wing Whirr. Abrupt bursts of 
sound produced by the wings are known in 
Sayornis, Pyrocephahis, two Ochthoeca spe- 
cies, three Muscisaxicola species, and prob- 
ably Xolmis ruhetra. Many other tyrannids, 
and many other passerines, at least rarely 
produce similar sounds. Wing Whirrs tend 
to occur in agonistic situations, sometimes 
in association with flight displays (e.g., in 
Tyranmis) or with Wing Raising (very 
regularly in M. maculirostris) . 

( 11) Other Wing, and Tail Displays. The 
\Ving Shuffle of Sayornis and Wing Flirt- 
ing of Pyrocephaliis appear to correspond 
closely. Probably Ochthoeca has similar 
displays, but related movements of the 
other genera may not be ritualized (except 
as a specialization for foraging in Musci- 
saxicola). Similarly, Tail Wagging in Say- 
ornis, Tail Flicking in Pyrocephalus, and a 
tail movement of Ochthoeca species cor- 
respond closely. Such a display is lacking 
in at least the more terrestrial genera, 
where it is to some degree replaced by 
Wing Raising. Tail movements known in 
the terrestrial species appear to be pri- 
marily balancing movements associated 
with locomotion. 

(12) Croicn Ruffled. Similar forms of 
Crown Ruffled are known in at least Say- 
ornis, Pyrocephalus, and Muscisaxicola. If 
there are other plumage displays, I have 
missed them by concentrating on sound 
recording and by not using binoculars suf- 
ficiently. 

In sum, this comparison shows that simi- 



larities among the displays of all these spe- 
cies are prominent, while widely different 
displays apparently do not occur. This is 
not to claim that there are no novel special- 
izations in the group; certainly the Head 
Row of Muscisaxicola is a novelty, special- 
ized to sliow the species-specific occipital 
patches, but it is not an elaborate inno- 
vation. Other less novel .specializations 
appear to be modifications of identifiable 
components, and do not observe generic 
limits. For instance, at least some species 
in the most open habitats appear to have 
lost the use of RRV in bouts while perched, 
and to have elaborated and stylized the 
Aerial Display instead. And at least the 
very terrestrial Muscisaxicola species ap- 
pear to have incorporated in the Aerial 
Display a Wing Raising which appears to 
be much rarer and less stylized in two 
Sayornis species. 

The displays, to the degree that they 
are now kno\vii, are less useful in delineat- 
ing the genera than are morphological and 
plumage characteristics. Some display char- 
acteristics may be of some help at this 
level, however. For instance, very brief 
vocalizations are typical only of Musci- 
saxicola (including M. hrevicauda, in 
\\]iich the component elements of displays 
are very brief); Xolmis pyrope may be the 
n^ost comparable species from this aspect 
in another genus. Further, the IPV and 
RRV are most similar in form and usage in 
the fumigata species-group of Xolmis, in 
Ochthoeca, and in Sayornis, which also 
intc^rgrade in plumage pattern and color- 
ation, and in habitat choice and many 
general behavioral features. 

Displays vary considerably within each 
genus, but mostly in ways already known 
in other genera. For instance, within one 
or more of the genera described herein, in 
some species, either the IPV or the SV 
may be prominent to the virtual exclusion 
of the other, or both may be common in 
the same species. This is also the case in 
the genus Sayornis, and shifts of a similar 
sort (usually involving the RV and CV) 



Evolution of Ground Tyrants • Smith and Vuilleumier 267 



are known in tlie phylogenetically distant 
tyrannic! genus Tijrannus (Smith, 1966). 

The display behavior of Muscisoxicola 
hrevicauda is at the moment the most 
obvious case of intrageneric distinctness, 
and the species is undoubtedly not fully 
typical of the genus Muscisaxicola. For a 
bird which is aberrant in so many ways, 
however, it is the similarities of its displays 
with those of other Muscisaxicola species 
that are striking, not the differences. Fur- 
ther, the displays of M. hremcauda are 
more like the known displays of Musci- 
saxicola species than they are like the 
known displays of any other passerines of 
the Andean chain. Considered alone, these 
displays suggest that M. hrevicauda is at 
least closely related to the members of 
Muscisaxicola, and differs largely through 
its emphasis on the LHV and related dis- 
plays. Considered in the light of ecology 
and other behavior (see above), plumage, 
morphology, and geographic distribution 
(see Chapter 1 by Vuilleumier), they help 
suggest that the species belongs in Musci- 
saxicola, as its most aberrant member. 

Finally, the behavioral similarities among 
tlie birds of these genera indicate that they 
belong to a natural group, the limits of 
which have not yet been determined. 
Other tyrannid genera which overlap with 
some in their habitats (e.g., Anairetes, 
Elaenia, and Mecocerculus in the f)aramo 
of Ecuador and Peru) are quite different 
in many aspects of their repertoires (Smith, 
in preparation). The various Tyrannus 
species have little in common with these 
species in the detailed forms of their dis- 
plays, even though there is at least general 
comparability in the usages (as should be 
expected for species with basically similar 
social behavior). In addition, sympatric 
nontyrannid passerines are generally quite 
dissimilar from the genera described herein 
in the forms of their displays, and this is 
true even of the superficially similar flight 
displays of some species of Pliryf!,ilus and 
Anthus. Thus there is little reason to 
suspect that the many similarities among 



the displays of the species of Muscisaxi- 
cola, Agriornis, Xolmis, Ochthoeca, Pyro- 
cephalus, and Sayornis are wholly or 
largely due to convergence in the face of 
common environmental pressures. The 
similarities of display behavior, like the 
similarities of nondisplay behavior, suggest 
close relationship. 

CONCLUSIONS 

General behavior patterns of the ter- 
restrial tyrannids in the genera Muscisaxi- 
cola, Agriornis, and Xolmis are similar in 
many respects, and no strongly divergent 
behavioral trends are known. In habitat 
preferences, foraging methods, patterns of 
locomotion, selection of nest sites and in 
the structures of the nests they build, they 
appear to vary along continua that trans- 
cend generic limits and relate these species 
to the less terrestrial members of Xolmis, 
Ochthoeca, Pyrocephalus, and Sayornis, all 
of which they also resemble in structure, 
coloration, and plumage patterns. Detailed 
similarities in the form and usages of the 
displays of members of all of these genera 
do not appear to be determined primarily 
by evolutionary pressures related to the 
habitats of the birds, and indicate that they 
belong to one phylogenetic group, distinct 
from many sympatric tyrannids. Some spe- 
cialized differences in display forms have 
developed among the genera, but none 
constitutes a major innovation. The forms 
of the displays give some clues as to re- 
lationships within the group, indicating 
that the less terrestrial genera may be more 
closely related to each other than any is to 
Muscisaxicola, and that Muscisaxicola is a 
fairly coherent genus (with the partial 
exception of the aberrant M. hrevicauda) 
that is probably closely related to Xolmis 
through X. pyrope. Greater detail of inter- 
relationships can likely be revealed by 
more detailed studies of display behavior, 
but it is not clear that this will necessarily 
aid significantly in the recognition of 
generic limits within this group. 



268 Bulletin Museum of Comparative Zoology, Vol. 141, No. 5 



LITERATURE CITED 

Cade, T. J. 1962. Wing movements, hunting and 
displays of the Northern Shrike. Wilson Bull., 
74: 386-408. 

GooDALL, J. D., A. W. Johnson, and R. A. 
Philippi B. 1957. Las aves de Chile, su 
conocimiento y sus costumbres. 1. Buenos 
Aires: Piatt Establecimientos Graficos S. A., 
441 pp. 

Hailman, J. P. 1960. A field study of the Mock- 
ingbird's wing-flashing behavior and its as- 
sociation with foraging. Wilson Bull., 72: 
346-357. 

Haverschmidt, F. 1953. Wing-flashing of the 
Graceful Mockingbird, Mimus gilvus. Wil- 
son Bull., 65: 52. 

Hellmayr, C. E. 1927. Catalogue of birds of 
the Americas and tlie adjacent islands in 
Field Museum of Natural History. Part V. 
Tyrannidae. Field Mus. Nat. Hist., Publ. 242, 
Zool. Ser., 13: 1-517. 

Hltdson, W. H. 1920. Birds of La Plata. Vol. 1. 
New York: E. P. Dutton, 244 pp. 

Ihering, H. von. 1904. The biology of the 
Tyrannidae with respect to tlieir systematic 
arrangement. Auk, 21: 313-322. 

Johnson, N. K. 1963. Biosystematics of sibling 
species of flycatchers in the Empidonax ham- 
mondi-oherhoheri-iorightii complex. Univ. 
Calif. Publ. Zool., 66: 79-238. 

KoEPCKE, M. 1954. Corte ecologico transversal 
en los Andes del Peru central con especial 
consideracion de las aves. Parte 1: Costa, 
vertientes occidentals y region altoandina. 
Mem. Mus. Hist. Nat. "Javier Prado," 3: 
1-119. 

. 1958. Die Vcigel des Waldes von Zarate. 

Bonn. Zool. Beitr., 9: 130-193. 

1964. Las aves del Departamento de 



Lima. Lima: Morsom S. A., 128 pp. 

Kunkel, p. 1962. Zur Verbreitung des Hiipfens 
und Lauf ens unter Sperlingsvogeln ( Passeres ) . 
Z. TierpsychoL, 19: 417-439. 

Lanyon, W. E. 1960. The Middle American 
populations of the Crested Flycatcher Myi- 
archus tyrannidus. Condor, 62: 341-350. 

. 1961. Specific limits and distribution of 

Ash-throated and Nutting Flycatchers. Con- 
dor, 63: 421-449. 

. 1963. Experiments on species discrimi- 
nation in Myiarchus flycatchers. Amer. Mus. 
Novitates, No. 2126: 1-16. 

. 1965. Specific limits of tlie Yucatan Fly- 
catcher, Myiarchus yiicatanensis. Amer. Mus. 
Novitates, No. 2229: 1-12. 

. 1967. Revision and probable evolution 



of the Myiarchus flycatchers of tlie West 
Indies. Bull. Amer. Mus. Natur. Hist., 136: 
329-370. 

Maclean, G. L. 1969. The nest and eggs of the 
Chocolate Tyrant Neoxolmis rufiventris 
(Vieillot). Auk, 86: 144-145. 

Marchant, S. 1960. The breeding of some S. 
W. Ecuadorian birds. Ibis, 102: 349-382. 

Moynihan, M. 1960. Some adaptations which 
help to promote gregariousness. Proc. 12th 
Internat. Ornithol. Congr., Vol. II: 52.3-541. 

. 1968. Social mimicry; character con- 
vergence versus character displacement. Evo- 
lution, 22: 315-331. 

Mumford, R. E. 1964. The breeding biology of 
the Acadian Flycatcher. Univ. Mich. Mus. 
Zool. Misc. Publ., 125: 1-50. 

Olrog, C. C. 1959. Las Aves Argentinas. Tucu- 
man, Argentina: Instituto Miguel Lillo, 377 
pp. 

Pena, L. E. 1961. Results of research in the 
Antofagasta Ranges of Chile and Bolivia. 1. 
Birds. Postilla, 49: 1-42. 

Peters, J. L. 1923. Notes on some summer birds 
of northern Patagonia. Bull. Mus. Comp. 
Zool., 65: 275-337. 

Schauensee, R. M. de. 1964. The Birds of 
Colombia and Adjacent Areas of South and 
Central America. Narberth, Pennsylvania: 
Livingston. 430 pp. 

Selander, R. K., and D. K. Hunter. 1960. On 
the function of wing-flashing in mockingbirds. 
Wilson Bull., 72: 340-345. 

Smith, W. J. 1966. Communication and relation- 
ships in the genus Tyrannus. Publ. Nuttall 
Ornitliol. Club, No. 6: 1-250. 

. 1967. Displays of the Vermilion Fly- 
catcher (Pyrocephalus rubinus). Condor, 69: 
601-605. 

. 1969. Displays of Sayornis phoebe. (Aves: 

Tyrannidae). Behaviour, 33: 283-322. 

■ . (in press, 1). Song-like displays in 

Sayornis species. Behaviour. 

(in press, 2). Displays and message 



assortment in Sayornis species. Behaviour. 

Stein, R. C. 1958. The behavioral, ecological and 
morphological characteristics of two popu- 
lations of tlie alder flycatcher, Empidonax 
traillii (Audubon). N. Y. State Mus. Sci. 
Serv. Bull., 371: 1-63. 

Todd, W. E. C, and M. A. Carriker, Jr. 1922. 
The birds of the Santa Marta region of Co- 
lombia: a study in altitudinal distribution. 
Ann. Carnegie Mus., 14: 1-611. 

Wetmore, a. 1926. Observations on the birds 
of Argentina, Paraguay, Uruguay, and Cliilc. 
Bull. U. S. Nat. Mus., 133: 1-448. 



msmmimmmmMmu 



Museum of 

Comparative 

Zoology 



iiiiliii 



Systematics and Natural History of the 

Mygalomorph Spider Genus 

Antrodiaetus and Related Genera 

(Araneae: Antrodiaetidae] 



FREDERICK A. COYLE 



HARVARD UNIVERSITY 

CAMBRIDGE, MASSACHUSETTS, U.S.A. 



VOLUME 141, NUMBER 6 
13 JULY 1971 



PUBLICATIONS ISSUED 

OR DISTRIBUTED BY THE 

MUSEUM OF COMPARATIVE ZOOLOGY 

HARVARD UNIVERSITY 

Bulletin 1863- 

Breviora 1952- 

Memoirs 1864-1938 

JoHNSONiA, Department of Mollusks, 1941- 

OccAsiONAL Papers on Mollusks, 1945- 



Other Publications. 

Bigelow, H. B., and W. C. Schroeder, 1953. Fishes of the Gulf of Maine. 
Reprint, $6.50 cloth. 

Brues, C. T., A. L. Melander, and F. M. Carpenter, 1954. Classification of 
Insects. $9.00 cloth. 

Creighton, W. S., 1950. The Ants of North America. Reprint, $10.00 cloth. 

Lyman, C. P., and A. R. Dawe (eds.), 1960. Symposium on Natural Mam- 
malian Hibernation. $3.00 paper, $4.50 cloth. 

Peters' Check-list of Birds of the World, vols. 2-7, 9, 10, 12-15. (Price list 
on request. ) 

Turner, R. D., 1966. A Survey and Illustrated Catalogue of the Teredinidae 
( Mollusca: Bivalvia ) . $8.00 cloth. 

Whittington, H. B., and W. D. I. Rolfe (eds.), 1963. Phylogeny and Evolution 
of Crustacea. $6.75 cloth. 

Proceedings of the New England Zoological Club 1899-1948. ( Complete sets 
only. ) 

Publications of the Boston Society of Natural History. 



Authors preparing manuscripts for the Bulletin of the Museum of Comparative 
Zoology or Breviora should send for the current Information and Instruction Sheet, 
available from Mrs. Penelope Lasnik, Editor, Publications Office, Museum of 
Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138, 
U.S.A. 



© The President and Fellows of Harvard College 1971. 



SYSTEMATICS AND NATURAL HISTORY OF THE MYGALOMORPH 
SPIDER GENUS ANTRODIAETUS AND RELATED GENERA 
(ARANEAE: ANTRODIAETIDAE) 



FREDERICK A. COYLE* 



CONTENTS 

INTRODUCTION 270 

ACKNOWLEDGMENTS 270 

NATURAL HISTORY 271 

Antrodiaetiis unicolor 271 

Antrodiaetus rohustus 284 

Antrodioeius pacificiis 285 

Antrodiaetus occultiis 288 

Antrodiaetus pugnax 288 

Antrodiaetus montanus 289 

Antrodiaetus hageni 291 

Antrodiaetus cerherus 292 

Antrodiaetus tjesoensis 292 

Antrodiaetus lincolnianus 292 

Antrodiaetus stygius 293 

Antrodiaetus apachecus 294 

Antrodiaetus roretzi 294 

Atypoides riversi 296 

Atypoides gertschi 300 

Atypoides hadros _ ^— - 302 

Aliatypus californicus 303 

Aliatypus no. 1 304 

Aliatypus no. 2 305 

Aliatypus spp. 306 

Summary and Discussion 306 

Natiu-al History Tables 312 

Natural History Figiues 316 

TAXONOMIC METHODS 325 

Analysis of Variation 325 

Material 328 

Collecting Methods —- 328 

Morphological Terminology — 328 

Methods of Presentation 329 



* Department of Biology, Western Carolina 
varsity, Cullowhee, North Carolina 28723. 

Bull. Mus. 



Uni- 



TAXONOMY 330 

Family Antrodiaetidae 330 

Key to Genera of Antrodiaetidae 331 

Genus Antrodiaetus 331 

Key to Species of Antrodiaetus 332 

The Unicolor Group 334 

The Lincolnianus Group 366 

The Roretzi Group 370 

Genus Aliatypus 372 

Taxonomic Tables -.- 373 

Taxonomic Maps 380 

. Taxonomic Figures 381 

EVOLUTION 393 

Phylogeny 393 

Dispersal Ability and Barriers 396 

Geographic Variation 396 

Speciation: Probable Historical Events 398 

Reproductive Isolating Mechanisms 400 



LITERATURE CITED 



ABSTRACT 



400 



A taxonomic section includes a characterization 
of the family Antrodiaetidae, a key to its three 
genera {Antrodiaetus, Atypoides, and Aliatypus), 
a revision of Antrodiaetus, and a preliminary study 
of Aliatypus. The family, except for two Japanese 
species of Antrodiaetus, occurs only in North 
America. Of the 13 species of Antrodiaetus, four 
are newly described: A. occultus, A. cerherus, A. 
stygius, and A. apachecus. Six new synonymies 
are recognized. Natural history data are reported 
for all species of the three genera. Antrodiaetid 
evolution, with emphasis on geographic variation 
and speciation in Antrodiaetus and Atypoides, is 
discussed. 



Conip. ZooL, 141(6): 269-402, July, 1971 



269 



270 Bulletin Museum of Cotnparative Zoology, Vol. 141, No. 6 



INTRODUCTION 

In general morphology and behavior 
AntrocUaefm (Figs. 109-112) and the 
closely related genns, Atypoicles (Coyle, 
1968), resemble the stout-bodied burrow- 
ing trapdoor spiders of the distantly related 
mygalomorph family Ctenizidae. All spe- 
cies construct tubular silk-lined burrows in 
the ground and prey nocturnally upon 
invertebrates which wander within reach 
of the burrow entrance. This entrance is 
well camouflaged and is closed off by a 
collapsible silken collar (or a rigid turret). 
The long-legged adult males abandon their 
burrows during the mating season and 
wander nocturnally in search of females. 
Most species are found in humid well- 
drained soil in forest or woodland habitats. 

Antrodiaetu.s and Atypoides together 
form a distinct monophylctic unit within 
the atypoid mygalomorph spider family 
Antrodiaetidae. The 13 species of An- 
trodiaetus and the three of Atypoides ex- 
hibit in toto a markedly disjunct geographic 
distribution with species in Japan, western 
North America, and eastern North America. 
The center of species diversity in both 
genera is in far western North America. 
These are clearly the dominant (most 
abundant and widespread) mygalomoqoh 
spiders above 35° latitude on this conti- 
nent. 

Taxonomic revisions of Antrodiaetus are 
nonexistant. Many species descriptions, par- 
ticularly pre- 1900 descriptions, are based 
on immature specimens or tiny samples, 
lack diagnostic characters, and are poorly 
illustrated. Misidentifications are common 
in the literature. Little has been written 
about the natural history of antrodiaetid 
taxa. Such a state of affairs is largely due 
to the relatively simple external morphol- 
ogy and covert behavior of these spiders. 

I have based the taxonomic revision of 
Antrodiaetus, like that of Atypoides (Coyle, 
1968), upon a thorough and objective 
search for diagnostically useful characters, 
my primary tool being an analysis of vari- 



ation of quantitative and qualitative mor- 
phological characters. Because I believe 
that many kinds of nonmorphological data 
provide useful, often vital, and interesting 
information about populations, taxa, and 
their past history, I have also made a start 
at a comparative behavioral, ecological, 
and life history study of the Antrodiaetidae. 
Using both morphological and nonmorpho- 
logical data, I have discussed the evolution 
of Antrodiaetus and Atypoides. Included 
is a preliminary consideration of the re- 
lationship of the only other antrodiaetid 
genus, AUatypus, to these two genera. It is 
my wish that this paper will stimulate fur- 
ther interest and continued research on 
these and related spiders. 

ACKNOWLEDGMENTS 

This work is a revised version of a doc- 
toral thesis completed at Harvard Univer- 
sity (Coyle, 1969). I am indebted to my 
advisor. Dr. Herbert W. Levi, for his en- 
couragement and generous support. Like- 
wise I am grateful to Dr. Willis J. Gertsch 
for his willingness to have me study these 
spiders which have deeply interested him 
for many years. Much of the material on 
which this study is based has accumulated 
in the American Museimi of Natural His- 
tory collection through his efforts. Dr. J. 
A. Beatty has shown uncommon interest 
in this study, generously supplying speci- 
mens and field observations. Dr. W. H. 
Bossert wrote the computer program for 
the analysis of variation. I am most deeply 
grateful to my wife, Judy, for moral sup- 
port, for assistance in field work, and for 
help with several tasks, especially typing. 

I sincerely thank the numerous other 
helpful people and institutions that have 
loaned material for study; these are named 
in the Methods section. C. W. Sabrosky and 
Drs. H. E. Evans, W. A. Shear, and A. A. 
Weaver have provided identifications of 
parasites, predators, and some prey. 

National Science Foundation Graduate 
Fellowships and a Richmond Fellowship 



Systematics and Biology of Antrodiaetus 



Coyle 



271 



from Harvard University have been the 
major sources of support for this study. 
Funds from NSF grant GB 3167 (Reed C. 
Rolhns, Harvard University, Principal In- 
\'estigator) helped support my field work 
during the smnmers of 1966 and 1967. 
Public Health Service Research Grant AI- 
01944 to Dr. Levi has helped to defray 
some expenses. 

NATURAL HISTORY 

Only three species of Antrodiaetidae — 
Antrodiaetus unicolor, Antrodiaetus roretzi, 
and Atypoides riversi — have had more than 
a few sentences written about their ecology, 
life history, or behavior. In this section I 
shall summarize the natural history litera- 
ture of the family and add original data. 
Primarilv because of mv field studies in 

1966, we now know much more about A. 
iinicolor than any other antrodiaetid. A 
comparative summary and discussion of 
the natural history of the family is placed 
at the end of this section after the separate 
consideration of each species. An enormous 
amount of information is still needed be- 
fore a valuable comparative biological 
study of the family can be completed. 
Hopefully this paper will trigger the search 
for such information. 

My observations were made chiefly in 
the field during the summers of 1966 and 

1967, but I also observed Hving spiders in 
the laboratory. Most observations oi> Alia- 
typus species were contributed by Wendell 
Icenogle. All specimens are numbered and 
have been placed in the American Museum 
of Natural History (AMNH) and the 
Museum of Comparative Zoology (MCZ); 
the correspondingly numbered field notes 
have been placed in the MCZ. Data on 
burrow architecture refer to adult female 
burrows unless stated otherwise. In the 
text many localities are represented by 
italicized letters; these localities are identi- 
fied in the records section at the end of the 
appropriate taxonomic species description. 



Antrodiaetus unicolor (Hentz) 
Ecology 

Geographic distribution and elevation 
range. Centered in the central and south- 
em Appalachian region of the eastern U. S. 
with peripheral populations as far west as 
the Ozark region and south to near the 
Gulf coast ( Map 1 ) . 0-6600 ft ( 0-2000 m ) . 

Habitat. Optimum conditions for A. 
unicolor are found in hnmid, cool, densely 
forested habitats with sandy loam soil. 
Such habitats are abundant and nearly 
continuous in the dissected terrain of the 
central and southern Appalachian Moun- 
tains and theii" foothills and westward over 
the Appalachian Plateau. The densest and 
largest burrow aggregations are usually 
found on the inclined surfaces of slopes 
and banks along ravines, streams, roads, 
and trails; at the base of rock (particularly 
sandstone) outcrops; and in other sheltered 
spots. Hemlock (Tsuga) and Rhododen- 
dron are often good plant indicators of 
optimum A. unicolor habitats. Over the 
more peripheral portions of the range of 
A. unicolor, one finds only scattered pock- 
ets of favorable habitat separated by larger 
areas of more or less unfavorable (dry, 
poorly drained, etc.) habitats. Although 
sandy loam is optimum, burrows also occur 
in other soil types, even sometimes in clay 
soil, decaying logs, etc. Bunows are nor- 
mally uncommon in rocky, shallow soil 
that is too well drained. 

Within favorable habitats burrows tend 
to be aggregated where the soil is con- 
stantly humid (at least around the lower 
portion of the burrow) and protected from 
erosion and flooding. In dry regions bur- 
rows are often clmnped only near springs, 
along stream banks, or near seeps at the 
bases of rock outcrops. Burrow aggre- 
gations are seldom found on near-hori- 
zontal ground, unless such ground is well 
sheltered under something such as a rock 
outcrop, a sandstone shelter cave, or the 
base of a large tree. Two limiting factors 
that may contribute to this clumping under 



272 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



shelters are the temporary localized flood- 
ing and the buildup of fallen leaf cover on 
open noninclined ground surfaces. Such 
flooding and (usually) leaf litter accumu- 
lation do not occur on the steeply inclined 
slopes and banks where A. unicolor bur- 
rows are common. On these inclined sur- 
faces, burrows are clearly aggregated in 
stable spots where erosion is minimal, e.g., 
moss covered areas and bare parts of the 
bank sheltered under the overhang of 
ground surface vegetation, exposed rocks, 
logs, large roots, and tree bases. Branson 
and Batch's ( 196(S ) statement that A. uni- 
color "lives in burrows mostly beneath 
fallen logs and deep litter" is erroneous. 

Population density. The densest and 
largest populations were found at C, D, I, 
J, K, O, P, and Q within the centi-al and 
southern Appalachian region. Even at such 
favorable localities, the i^opulation con- 
sisted of dense aggregations in optimum 
habitat spots and scattered individuals in 
intervening areas of less suitable habitat. 
One of the densest such burrow aggre- 
gations covered an area of about 50 m^ and 
contained an estimated 80 to 100 adult 
females and several thousand immatures. 
Small dense aggregations were common 
within the larger aggregations in the above 
localities. A 0.1-m- area at O contained 
four brooding females and 25 immatures. 
An 0.81-m- area at P contained three adult 
females and 198 immatures. A 225-cm- 
area at B contained two adult females and 
31 immatures. In many dissected regions 
the populations are concentrated and es- 
sentially continuous throughout the inter- 
connecting ravines, young valleys, and 
coves, with the surrounding matrix of dryer 
ridges and upland areas very sparsely 
populated. In the peripheral portion of 
the A. tmicolor range, aggregations are 
(like the favorable habitat) discontinuous 
and generally smaller and less dense. 

Associated bumncinfi, myfialomorph spi- 
ders. Throughout most of its range, A. 
unicolor is clearly the most abundant 
mygalomorph spider. Except for areas of 



overlap with the northern Antrodiacttis 
rohustus (Map 1), only in southern and 
western parts of its range (in lowland 
liabitats and in the Ozark region) is the 
abundance of A. unicolor equaled and 
sometimes surpassed by other burrowing 
mygalomorph species. I have found the 
follo\\'ing such spiders living side by side 
with A. unicolor at the following localities: 
Afypoides hadros Coyle at F (These two 
species have been found together in other 
southern Illinois localities by other collec- 
tors.); MyrmekiuphiJa fluviatilis (Hentz) 
at H, O, R, Q, and Mammoth Cave, Ken- 
tucky; Cyclocosmia triincata (Hentz) at 
Q, R, and David Crockett State Park, 
Tennessee; Ummidia audouini (Lucas) at 
Chickasaw State Park, Tenn.; Ummidia 
carahivora (Atkinson) at N; and Ummidia 
spp. at Q and Blanchard Springs, Arkansas. 
At F, A. hadros was common but not as 
abundant as A. unicolor. At both Q and R, 
M. fluviatilis was common but much less 
abundant than A. unicolor. In one low 
road bank at Q, A. unicolor, M. fluviatilis, 
C. truncata, and Ummidia sp. lived to- 
gether, the latter two species being rare. 
At all other localities cited above, the other 
mygalomorph spiders were very rare. 
Atypus species are sympatric with A. uni- 
color but uncommon; I have never ob- 
served them at the same locality. I have 
not found Lycosidae burrows in A. unicolor 
aggregations. 

Life History 

Males. Males of A. unicolor (and all 
other Antrodiaetidae) closely resemble fe- 
males in external anatomy and behavior 
until the striking metamorphosis of the 
final male ecdysis. Only a few obvious, 
sexually dimorphic characters — a sclero- 
tized lip at the male genital opening, 
swollen male pedipalpal tarsi, and well- 
developed seminal receptacles of the fe- 
males — appear during the later immature 
instars. The penultimate male undergoes 
the final molt ^^ithin its burrow, and the 
adult remains there without feeding until 



Systematics and Biology of Antrodiaetus 



Coyle 



273 



certain factors trigger its abandonment of 
the burrow. Upon emergence, it wanders 
over the ground surface in search of female 
burrows, mates, and eventually dies. 

The data summarized in Figure 1 in- 
dicate that the period of male wandering 
and mating for the entire species popu- 
lation of A. unicolor lasts from late July 
through late December. Because of oc- 
casional long-lived males, the mating sea- 
son may be shorter than the period during 
which wandering males have been col- 
lected. The high frequency of wandering 
male records during October indicates that 
this may be the chief mating season of A. 
unicolor. However, all August records of 
mature males collected within their bur- 
rows were from cooler climate localities 
(C, I, and P) than the October collections 
and therefore possibly belong to earlier 
(September ?) emerging populations. 

Evidence indicates that within any given 
local population the time of male emer- 
gence (and probably mating) occurs more 
nearly synchronously and is probably trig- 
gered by local climatic conditions. At N, 
J. Beny conducted a systematic, year-long 
pitfall trap collection resulting in a rel- 
atively complete record of male emergence 
in that population (J. Beatty, personal 
communication). Traps were put out for 
approximately 24-hour periods iwo times 
each month (only once a month in the 
winter) in each of 11 different areas. Four- 
teen of the 18 adult males collected. were 
taken from 22 to 24 October (ten of these 
on 24 Oct. ) following the first heavy rain- 
fall after a dry summer. The other four 
males were collected on 1 October, 7 No- 
vember, and 3 January. Pitfall traps main- 
tained continuously from 7 October to 10 
November at E by J. Nelson several years 
later (J. Beatty, personal communication) 
yielded 11 males during the initial week 
(7-14 Oct.), two the second week, one the 
tliird, one the fourth, and none during the 
final week (3-10 Nov.). A similar pitfall 
collection maintained by J. Nelson the 
following year at G from 10 September to 



2 November yielded the first male during 
the week 28 September to 5 October, seven 
males the next week, one the following 
week, none the next week, and one during 
the final week (26 Oct.-2 Nov.). Collections 
during August revealed a pre-emergence 
synchrony of adult males at localities 
C, I, and P. Twenty-one of the 22 
adult males collected were within their 
own burrows. I searched hard for wander- 
ing adult males during day and night but 
found only one. It appears that very few 
males had emerged and that proper con- 
ditions could have set off a large synchro- 
nous emergence. 

A few winter records exist for wandering 
males of A. unicolor — an early January 
record from N and a literature record from 
D of "males observed on [the] surface 
during late winter" (Branson and Batch, 
1968). I have not seen the latter males 
(from D) so only assume that they are 
A. unicolor. Such records are probably 
either aberrant late emerging individuals 
or indix'iduals which have extended their 
above-ground survival time in particularly 
protective microhabitats. 

Females. Potfern of carlii postemhnjonic 
development. Enough data were obtained 
from stereomicroscopic examination of each 
preserved A. unicolor brood to establish 
the following pattern of early postembry- 
onic development. Hatching involves the 
simultaneous (or nearly so) shedding of 
the chorion and embryonic or "prelarval" 
(Vachon, 1958) cuticle. Tlie resulting 
instar, which I shall call the first instar, is 
the "first free postembrvonic stage" of 
Holm (1954) or "larva" of Vachon (1958). 
The second instar corresponds to the "first 
complete postembrvonic stage" of Holm 
(1954) or "first nymph" of Vachon (1958). 

The segmentation and appendages of the 
late embryo, as well as the dark egg teeth, 
are visible through the chorion. During 
hatching, first the chorion splits anteriorly 
and peels back over the abdomen. Usually 
the embr>'onic cuticle splits very soon after 
the chorion and also peels posteriorly, and 



274 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



the chorion and cuticle are shed simul- 
taneously from the posterior of the abdo- 
men and remain attached to each other. 
Occasionally, possibly because the interval 
between the splitting of these two mem- 
branes increases, they are shed separately. 
These conclusions about the timing of 
hatching are based on the following ob- 
servations of preserved material: 1) Rel- 
atively few spiderlings in any hatching 
brood were actually in the process of peel- 
ing back or shedding the chorion or em- 
bryonic cuticle, and, in the great majority 
of those that were, the chorion was either 
peeling back or was still attached to the 
spiderling while the embiyonic cuticle was 
jDeeling back. 2) Almost all shed chorions 
observed contained the embryonic cuticle. 

The embryonic cuticle is very thin and 
transparent, lacks setae, has blunt non- 
functional chelicerae, and possesses a dark 
egg tooth at the base of each pedipalp. 
The first instar (Fig. 4) possesses several 
tarsal setae, lacks a rastellum or fang on 
the chelicera, lacks spinneret spigots, has 
anterior lateral spinnerets, and lacks pig- 
mentation, except for that \\'hich slowly 
de\'elops around the eyes. Tlie second 
instar (Fig. 5) has functional chelicerae 
and spinnerets and can locomote rapidly. 
Tlie chelicerae possess a few rastellar 
macrosetae and fully formed fangs, the an- 
terior lateral spinnerets are lost, the other 
two pairs of spinnerets possess spigots, 
setae are plentiful, and the body is pig- 
mented. 

This pattern of development is the same 
as that described for Ischnothcle karschi 
(Dipluridae) by Holm (1954) and for 
Atypus karschi (Atypidae) and Hepia- 
thela kinmrai (Liphistiidae) by Yoshikura 
(1955, 1958). It is considered by Holm 
( 1954 ) to be the most primitive pattern in 
spiders. Vachon ( 1958 ) reviewed these 
studies and is in agreement on the simi- 
larity of patterns in the above three species. 

Timing of brood development . The nu- 
merous field collections of A. unicolor 
broods yield substantial data on the timing 



of brood development in nature within the 
species population as a whole. As shown in 
Figure 2, the egg stage is found from late 
June to mid-August, the first instar from 
mid-July to mid-September, and the second 
instar from mid-August through at least 
December. The relatively small number of 
broods in the process of hatching and 
molting is apparently a result both of 
the brevity of hatching and ecdysis in in- 
dividual spiderlings and of the close syn- 
chrony of development within each 
individual brood. The regular temporal 
pattern of the brood stages and the absence 
of more than a single brood in the burrow 
of any brooding female strongly indicate 
that A. unicolor produces only one brood 
per year in nature. As Table 1 shows, 
brood development is not strictly syn- 
chronous within a local population, popu- 
lations C and O showing the largest 
amount of variation. These data also indi- 
cate little or no lag in timing of brood 
development at warm climate localities 
(N, O, Q, R) behind that in cooler lo- 
calities. 

Substantial evidence indicates that the 
spiderlings overwinter in the second instar 
within the maternal burrow and disperse 
in the spring. As indicated in Table 1, 15 
brooding females were collected between 
14 September and 3 December from three 
localities. Each brood consisted entirely 
of second instar spiderlings, and none of 
the broods appeared to be diminished. In 
the December collection from Q, the 
spiderlings were quite sluggish as a result 
of the cold weather. During this collection 
and the September collection at F, I 
searched hard for bun-ows constiiicted 
by second instar spiderlings but the 
smallest burrows found contained only later 
instars. Two samples of second instar spider- 
lings from the December Q broods were 
kept alive at a nearly constant temperature 
of 12°C. One sample was checked in mid- 
February and the other in mid-March and 
both were still 100 per cent second instar 
broods. These were then released at room 



Systematics and Biology of Antrodiaetus 



Coylc 



275 



temperature into a jar of sand, and they eon- 
structed burrows within 24 hours. 

Few data exist concerning longevity and 
the time normally required for an in- 
dividual to mature. Oviposition to the 
beginning of the third instar may take nine 
or ten months. I \\'Ould estimate at least 
three or four years from egg to sexual 
maturity. I have collected two brooding 
females (384 at Blanchard Springs, Ark., 
on 11 Sept.; 378 at F on 15 Sept.) with 
new cuticle visible under the old cuticle, 
thus indicating that females continue to 
molt after becoming sexually mature and 
after producing broods. The following 
evidence indicates that a given female may 
be capable of producing more than one 
brood. A number of brooding female bur- 
rows were each located in the center of a 
cluster of homogeneously-sized burrows of 
small immature spiders. Three of these 
brooding females (787 at Fort Mountain, 
Ga., 433 at Q, and 448 at F) were completely 
isolated from other large burrows. 

Brood size and egg, size. The sizes of 
complete broods collected at different lo- 
calities are given in Table 2. Brood size 
statistics for the species as a whole are 
listed in Table 3. All populations with 
significant sample sizes show a wide range 
of variation in brood size and overlap ^^'ith 
one another, but the means are often quite 
different. Table 4 indicates that there is 
much variation in egg size of A. tinicolor. 
Egg size variation within an individual 
brood is quite small. 

Behavior 

Burrow structure. A. unicolor constructs 
a roughly tubular silk-lined burrow which 
is widened just below the entrance and at 
the bottom end (Figs. 6-9). The detailed 
shape of the burrow, its size, and the thick- 
ness of its silk lining vary greatly, depend- 
ing on the size, sex, and reproductive state 
of the spider, and especially on soil con- 
dition. Tables 5 and 6 show the variation 
in burrow length and entrance diameter 



in A. unicolor. The latter measurement is 
more closely related (directly) to body 
size than is burrow length, which is more 
influenced by soil conditions. The mean 
size of penultimate male burrows is mark- 
edly smaller than that of adult female 
burrows, probably because of the smaller 
body size of the males. Soil with many 
pebbles, rocks, or roots usually contains 
crooked and /or shortened burrows. Bur- 
rows in hard clayey soil are shorter than 
those in finn sandy soil. Soil stability (and 
possibly humidity) appears to have con- 
siderable effect on the thickness of the 
burrow lining. Burrows in loose soil tend 
to have thicker linings than those of the 
same population in more stable soil. Usu- 
ally the upper part of a burrow has the 
thickest silk lining, this probably being, at 
least in part, an adaptive response to the 
relative instability (or low humidity) of 
surface soil. Tlie rest of the burrow is 
sometimes only thinly lined with silk. 
Brooding females (even with unemerged 
broods) often have thicker burrow linings 
than adult females without broods. All 
brooding females at O had markedly 
thicker burrow lining than many brooding 
females at other localities. 

Burrow inclination is strongly correlated 
^^'ith the inclination of the immediate 
ground surface. The burrow pitch — here 
defined as the angle fonned between the 
mean longitudinal axis of the burrow and 
the line originating at the entrance and 
perpendicular to the surrounding ground 
surface (i.e., the deviation of the burrow 
from the perpendicular) — of the great 
majority of A. unicolor adult female bur- 
rows is 0-15°. Exceptions are often the 
result of rock or root barriers in the soil. 
Only one or two burrows sloped upward 
from the entrance. Tlie probable adaptive 
significance of such a near-peipendicular 
burro\\' pitch is that it removes as much of 
the burrow as possible from the surface 
soil, which is structurally and climatically 
(humidity and temperature) more un- 
stable than the deeper soil layers. This 



276 Bulletin Museum of Comporative Zoology, Vol. 141, No. 6 



might be particularly important on the 
steeply inclined banks often inhabited 
by A. iinicolor where surface erosion is 
relatively great. Humidity, structural sta- 
bility, or temperature gradients are di- 
rectional cues that might be used in 
burrow construction. 

Occasionally, A. tinicolor burrows have 
single, short, dead-end side branches. 
These are either close to the surface (Fig. 
10) or, rarely, at the bottom end. One 
Y-shaped burrow (389 at F) with two 
entrances was found. Only once have I 
found a spider in a side branch. 

Entrance structure. The silk lining of 
the burrow is extended above the soil sur- 
face to form a flexible collar which is col- 
lapsed inward to close off the open end 
of the burrow (Figs. 11-14, 40-43, 59). 
This collar is composed of excavated soil 
particles and materials from the immediate 
ground surface (such as pieces of twigs, 
leaves, conifer needles, moss, or just sur- 
face soil) held together by silk so that its 
outer surface is camouflaged and its inner 
surface silk-lined. The kind and quantity 
of surface material incorporated into the 
collar depends on its abundance on the 
immediate ground surface. When open 
the collar is more or less erect, and when 
closed it is roughly flush with the ground 
surface. The relative height of the collar 
varies noticeably within populations. Popu- 
lation O appears to have collars with the 
largest height/diameter ratio. 

The spider collapses the collar in a 
bilateral manner by pulling inwardly on 
two opposing sectors (lateral to the 
spider), producing a condensation of fold- 
ing at the t\vo opposing points where these 
sectors meet. When the burro\\' entrance 
opens onto an inclined ground surface, 
these two points are at the ventral and 
dorsal sectors of the entrance opening, 
because of the spider's stereotyped forag- 
ing posture orientation. The collar (espe- 
cially of the larger spiders) is frequently 
somewhat reduced at these two points, 
which, if the entrance opening happens 



to be slightly elliptical, also correspond 
with the ends of the long axis of such an 
ellipse. Careless obsei"vation of such a 
collapsible collar by Atkinson (1886) led 
to his erroneous description: "Each door 
is a surface of a half circle, is hung by a 
semicircular hinge, and the two meet, 
when closed, in a straight line over the 
middle of the hole. ..." Unfortimately 
the majority of subsequent descriptions 
(Simon, 1892; Comstock, 1912 and 1940; 
Gertsch, 1949; Kaston and Kaston, 1953; 
Kaston, 1964 and 1966; Pimentel, 1967) of 
Antrodioettis bm^row entrances are ap- 
parently based on Atkinson's inaccurate 
description. 

Under certain conditions, a closed collar 
may be held shut by a small amount of 
silk spun over the inside surface of the 
collar. Such sealing is usually performed 
before a spider molts. Tlie burrow en- 
trances of brooding females are sometimes 
similarly sealed. Soil plugs are apparently 
used much more infrequently to seal the 
burrow entrance. I have found these soil 
plugs in only five medium-sized burrows 
(at O and /) and one adult female burrow 
(1115 at P). Tlie immature spiders were 
molting and the adult female had an egg 
sac. In each of these, the upper 0.5-1.5 
cm of burrow just behind the closed collar 
was packed tightly with soil (Fig. 8). 

Burrow construction hehavior. Nonnally 
an individual probably occupies and en- 
larges the same burrow during its entire 
life span, the burrow site initially being 
chosen by the second instar spiderling. 
However, I have collected several medium- 
sized immatures and one adult female A. 
unicolor on the ground surface away from 
any burrow and have occasionally observed 
empty burrows with open collars, no spider 
remains, and no penultimate male molts. 
Under laboratory conditions, large im- 
mature male and female spiders and adult 
females are capable of constructing bur- 
rows from scratch. These observations 
indicate that the abandonment of burrows 



Systematics and Biology of Antrodiaetus 



CoyJc 



Til 



and the founding of new burrows at differ- 
ent sites may not be uncommon in nature. 

The following account of burrow con- 
struction behavior is based on laboratory 
observations of ten large immature (male 
and female) and mature (female) spiders 
burrowing into humid packed sand; it is 
incomplete and subject to revision when 
more detailed observations are completed. 
The initial step is the construction of a 
shallow enclosure or cell. Next, sometimes 
after a period of inactivity, the spider ex- 
tends this closed cell deeper into the soil 
until the elongate burrow is formed. Most 
observations were made on the initial cell 
construction phase, ^^'hich lasted from 15 
to 30 minutes. 

Four rather well-defined behavioral 
components form cell construction be- 
havior: 1) burrowing, 2) excavating, 3) 
silk application, and 4) raking. Burrowing 
consists of pushing into the sand, forcing 
it apart, and compacting it. An initial de- 
pression is made by forcing the palps and 
first legs into the sand surface. Very soon, 
however, as the size of the depression in- 
creases, the other appendages enter into 
the burrowing movement. Full-formed 
burrowing behavior consists of the chelic- 
erae pressing forcibly against the side of 
the bottom of the depression (or burrow) 
with the legs flexed over the dorsum of 
the cephalothorax and abdomen and press- 
ing in the opposite direction against the 
opposing side of the cavity (Fig^ 56). 
Excavating involves digging sand from the 
bottom of the depression, transporting it, 
and releasing it on the soil surface. The 
chelicerae are elevated and spread apart 
laterally, and the fangs are extended. These 
are forced ventrally into the sand, flexed, 
and then lifted away with the sand held in 
between. Tlie spider then pivots to reverse 
its direction and moves up to the top of 
the cavit\^ The palps aid in holding the 
sand in the chelicerae. With its front legs, 
pedipalps, and chelicerae it reaches out of 
the cell opening over the soil surface (Fig. 
57). It then releases the load by opening 



its fangs and chelicerae, and pulls away 
from the load while holding the load 
against the soil surface with its pedipalps. 
Silk application includes circular and 
lateral movements of the abdomen and the 
entire body (with the spider facing down 
into the cell) and sweeping movements of 
the spinnerets (particularly the elongate 
posterior lateral spinnerets), during which 
silk is applied to the cell wall and most 
abundantly to the rim of the cell opening 
(Fig. 58). In raking, the first legs, the 
pedipalps, and sometimes the chelicerae 
and fangs reach out past the rim of the 
cell opening and pull surface detritus and 
soil back to the rim of the opening and 
partially over the opening. Tliis material 
is often compacted by being pressed 
against the chelicerae by the first legs and 
pedipalps. After the spider releases it, it is 
held in place by silk previously applied to 
the rim. 

Cell construction usually begins after 
considerable searching behavior in which 
wandering alternates with mild burrowing, 
^^'hich is usually released by depressions in 
humid sand. Eventually a series of bur- 
ro\\'ing mo\'ements is repeated in one spot. 
Burrowing usually continues until a fairly 
deep depression is formed, and then ex- 
cavating begins and alternates sporadically 
\\'ith burrowing. The frequency of bur- 
rowing decreases, and soon silk application 
begins and, sporadically, but with increas- 
ing frequency, accompanies burrowing and 
excavating. If it follows excavating, silk 
is usually applied to the sector of the open- 
ing rim and cell wall where the soil load 
^^'as dropped. When the cell becomes deep 
enough to contain the spider, raking begins 
and often follo^^'S excavating. Usually only 
a few instances of raking will suffice to 
pull the rim of the cell opening shut (or 
almost shut) and thus end the initial stage 
of burro\\' constiiiction. 

Burrowing, excavating, and silk appli- 
cation continue during the rest of burrow 
construction, but the manner in which the 
load of sand is deposited was quite differ- 



278 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



eiit in the only two spiders observed after 
initial cell construction. Instead of being 
deposited on the ground surface, each load 
\\as pressed into the burrow wall just be- 
low the top of the closed cell. Apparently 
the pedipalps held the sand against the 
\\'all while the chelicerae and fangs were 
pulled awav, then the sand was pushed 
into the wall with the rastellar area of the 
chelicerae. This behavior eventually caused 
the area around the entrance to bulge 
outward into a mound. Such internal dep- 
osition of soil, if it is normal behavior, 
would seem to be an effective adaptation 
for minimizing the exposure of the spider 
to predation. Collar formation has not 
been observed. 

Atkinson (1886) observed burrow con- 
struction behavior in several spiders kept 
in captivitv, and although his description 
is often difficult to understand, his ob- 
servations generally agree with those 
above. He observed too, that after the 
initial cell is closed over, the excavated 
soil is deposited and pushed into the wall 
within the upper end of the burrow. 
"Dome" apparently refers to the closure 
over the cell. Atkinson claimed that the 
spider would apply "viscid liquid" as well 
as silk to the excavated soil deposited at 
the rim of the cell opening. It is not clear 
what he meant by "viscid liquid." 

Egg sac structure and position. The egg 
mass of A. unicolor takes the form of a 
flattened sphere or clli]isoid with a maxi- 
mum dimension of usually 9-14 mm and a 
thickness of usually 5-8 mm. Apparently 
tliis mass is formed on a portion of the 
burrow lining which has been thickened 
with a few additional layers of silk. Several 
layers or sheets of silk are then apparently 
applied over the egg mass and hold it 
firmly against the burrow wall, each sheet 
larger and attached to the burrow wall 
slightly farther from the egg mass than 
the previous sheet (Figs. 6-9). J. R. Tripp, 
who has recently observed oviposition and 
egg sac construction in this species, con- 
firms that this is the method of egg sac 



construction ( personal communication ) . 
Egg sac silk has the same texture and white 
color as burrow lining silk. 

As shown in Tables 5 and 6, the position 
of the egg sac in the burrow is quite vari- 
able for A. unicolor. However, while some 
population samples exhibit wide variation 
in egg sac position, others (Q and espe- 
cially O) show much less variation and 
have all egg sacs deep within the burrow. 
In the populations in which brooding fe- 
males were often actively feeding (O, Q, 
and R), egg sacs were well below the sur- 
face. It is possible that these tsvo phenom- 
ena are adaptively linked, since it is likely 
that egg sacs near the entrance would be 
more easily damaged during prey capture 
than egg sacs attached deep \\'ithin the 
burrow. Possible environmental causes 
should be investigated. All but one of the 
101 egg sacs were placed on the ventral 
sector of the burrow wall. Most egg sacs 
produced a localized constriction of the 
bmrow lumen, but never so much that the 
spider could not maneuver past it. In at 
least most cases this constriction is counter- 
acted, because the burrow is somewhat 
widened in the vicinity of the egg sac. 

Overwintering. On 28 December bur- 
rows were observed at B on an overcast 
afternoon (noon temperature = 0°C). Five 
centimeters of snow lay on the ground 
except under the shelter of a sandstone 
wall where most of the burrows were lo- 
cated. The upper 1 cm of the sand soil was 
dry, loose, and unfrozen; the next 9-15 cm 
frozen; and below this humid, packed, but 
unfrozen. No burrows were seen with open 
collars. Close examination of about 20 
small and medium-sized burrows revealed 
only two entrances sealed inside with silk 
and none plugged with soil. Four burrows 
were completely excavated. Two of these 
had one or t\vo very thin silk septa spun 
across the narrowed central part of the 
burrows. Two of the burrows did not 
extend below the frozen zone, the other 
two did. All four spiders were quite slug- 



Systematics and Biology of Antrodiaetus • Coyle 



279 



gish but became active after several 
seconds in my hand. 

On the night of 2 December and in day- 
light the next morning, I observed a bur- 
row aggregation at Q. The ground was 
unfrozen, but the air temperature was 3°C 
at 9 P.M. and -2°C at 8 a.m. At night, 
during a light snow and rain, I was sur- 
prised to find that most burrows of all 
sizes were open, and most of these had 
spiders at the entrance in normal foraging 
posture. These spiders were more sluggish 
than usual, however. In the morning ( clear 
weather), a few entrances were still open 
but no spiders were in the entrances. No 
sealed entrances could be found. This 
ability to remain active at relatively low 
temperatures may contribute to the re- 
markable success of Antrodiaetus in the 
temperate zone. 

Bella vior of brood inp, females and 
broods. Tlie data summarized in Table 7 
show that the broods emerge from the egg 
sac just before, during, or just after they 
molt from the first to the second instar. 
In broods that were emerging when col- 
lected, only a few first instar spiderlings 
(when present) but most of the second 
instar spiderlings (when present) were 
found outside the egg sac. Also, in emerg- 
ing broods and in the few completely 
emerged broods still with egg sacs, most of 
the first instar exuviae were found in the 
egg sac. One exception, 477 at 7, had these 
exuviae scattered all over the burrow wall. 
These data indicate that most emergence 
is performed by the active second instar 
spiderlings. Tliere is no evidence indicat- 
ing how the spiderlings escape from the 
egg sac or how the egg sac is disposed of. 
In 59 of the 66 burrows with completely 
emerged broods, the egg sac was no longer 
present. 

Table 8 indicates that while in some 
populations (O, Q, R) brooding females 
with egg sac broods do not seal their bur- 
row entrances and usually continue forag- 
ing, in others (B, C, /) they commonly 
maintain sealed burrows. Such sealing of 



burrows is in contrast to neighboring con- 
temporary nonbrooding females, which, 
unless molting, usually kept their entrances 
unsealed and were actively feeding. How 
long and for what function such burrows 
are kept sealed is not known. I was unable 
to determine whether the sealing of bur- 
rows containing emerged second instar 
spiderlings was initially produced by the 
adult and thus signifies inactivity or 
whether it is simply the result of spider- 
lings trailing dragline silk over the inside 
of the entrance. At least a number of such 
brooding females had unsealed burrows 
and were actively feeding. 

First instar spiderlings, extremely slow 
moving, are sometimes seen moving within 
the egg sac. Second instar spiderlings are 
active — they can move quickly in response 
to stimuli, wander over the burrow wall, 
and are capable of performing many 
behavior patterns necessary for inde- 
pendent existence. Several second instar 
spiderlings released on humid sand in the 
laboratory constructed individual burrows 
and assumed normal foraging postures. 
Second instar spiderlings are capable of 
spinning draglines and nonnally appear 
to do so as they move about. When 
a brooding female burrow with a closed 
entrance is first discovered, second instar 
spiderlings are often found in the upper 
end of the burrow, but these always re- 
treat to the bottom of the burrow as it is 
excavated. This may be a photonegative 
response, but other behavioral controls 
must be operative in preventing these 
spiderlings from escaping at night if the 
burrow entrance is open and the brooding 
female is feeding (such as at Q in Sep- 
tember and December). Each of the five 
broods observed on 2-3 December at Q 
was aggregated in the bottom end of the 
burrow, several in dense clusters; and all 
spiderlings were sluggish. 

There are no data on whether spiderlings 
normally feed within the parental burrow 
before dispersal. An observation of a sec- 
ond instar spiderling which had not yet 



280 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



constructed a burrow but which carried 
a collembola in its cheHcerae, and obser- 
vations of other second instar spidcrhngs 
capturing prey after constructing their 
burrows, indicate that they are at least ca- 
pable of feeding before dispersal. Intra- 
brood cannibalism is possible, but the 
broods observed in fall and winter did not 
appear to be smaller on the average than 
those collected earlier. 

Moltinfi behavior. Fragmented remains 
of one and occasionally two old exuviae 
were often found to be bound with silk 
and compacted into the bottom end of 
burrows. I collected (in July, August, and 
September) ten immature spiders each 
\\'ith an incompletely sclerotized cuticle 
and a freshly fragmented exuvia mashed 
into the bottom of its burrow. Seven of 
these recently molted individuals had 
sealed their entrances with silk, and three 
were discovered by scraping the soil 
(whether the entrances were sealed could 
not be detennined ) . Four other immatures 
(also collected during the summer) were 
found just after ecdysis; thev were much 
less sclerotized and could not move 
quickly. These were at the bottom end of 
their burrows, and the intact exuviae were 
attached lightly to the ventral sector of 
the wall of each burrow at one-half to 
three-fourths of the distance from the en- 
trance to the bottom end. One exuvia was 
attached with its venter against the burrow 
wall, and its anterior end heading up the 
burrow. The orientations of the other 
three were not recorded. The entrances 
of three of these burrows ( 133 at /, 951 at 
C, and 747 at Brasstown Bald, Ga.) were 
sealed with silk, the fourth (355 at /) was 
filled with a soil plug. Another medium- 
sized immature spider (7040 at Q) was 
excavated shortly before ecdysis, which it 
underwent within a glass vial kept humid 
by a moist piece of paper towel. Ecdysis 
lasted ten minutes. Within four days of 
ecdysis, the spider had thoroughly frag- 
mented the exuvia and formed it into a 



compact mass bound together with a small 
amount of silk. 

Twenty-seven burrows of recently 
molted adult males were excavated. All 
burrow entrances were closed, but 23 were 
discovered by scraping the soil surface; 
whether they were sealed could not be 
determined. Three of the other four were 
sealed with silk; the fourth was unsealed. 
In all but two of these 27 burrows the 
penultimate exuvia was partly broken up, 
bound with a little silk, and packed lightly 
into the bottom end of the burrow. Some- 
times this exuvia and the end of the bur- 
row were covered with several thin sheets 
of silk. The other two burrows each con- 
tained a pale unsclerotized male resting 
motionless at the bottom end of the bur- 
row. In each the freshly shed exuvia was 
attached — venter against the wall, anterior 
end heading up the burrow — to the silk 
lining of the ventral sector of the wall, 
two-thirds (409 at F) and three-fourths 
(450 at Blanchard Springs, Ark.) of the 
distance from the entrance to the bottom 
end. 

Apparently, molting behavior in penulti- 
mate males and other immatures is similar. 
In summary, ecdysis usually takes place in 
a sealed burrow on the ventral sector of 
the burrow wall in the lower half of the 
burrow (but above the bottom end). 
Shortly after ecdysis, the spider moves to 
the bottom end. After sclerotization 
reaches a certain point, the spider detaches 
the exuvia from the burrow wall, breaks it 
up, and compacts it into the bottom end 
of the burrow, adding silk. Adult males 
do not usually fragment the molt as 
thoroughly or pack it into the end of the 
burrow as compactly as do females or im- 
mature males. Tlie sealing of burrows 
before molting may help protect the molt- 
ing spider from predation and may main- 
tain a higher, more constant humidity 
within the burrow. 

Defensive ])eh(ivi()r. When spiders in 
nomial foraging posture are disturbed by 
light, substrate vibration, or prodding with 



Systematics and Biology of Antrodiaetus 



Coijle 



281 



a stick, they sometimes retreat down the 
burrow without closing the collar or may 
quickly close the collar, often remaining 
for a time just below the collar holding it 
closed. During the later stages of excavat- 
ing to expose a burrow, one finds most 
adult female spiders at the bottom end of 
the burrow in an aggressive defensive pos- 
ture. The spider faces up the burrow with 
its cephalothorax reared backwards at an 
angle to the abdomen. Legs IV and III 
and possibly II hold onto the burrow wall 
to anchor the spider. The other legs and 
pedipalps are raised and spread apart. The 
chelicerae are also raised and spread apart 
laterally with the fangs extended. If 
prodded, the spider strikes forcibly with 
a sudden downward movement of the 
cephalothorax and chelicerae. Such a de- 
fensive posture within the confines of the 
bottom end of the burrow limits the ap- 
proach of an attacker to that part of the 
spider protected by its chelicerae. Younger 
individuals and adult males tend to be less 
aggressive and often do not exhibit such 
defensive behavior while being excavated. 
Medium and large spiders will often ex- 
hibit similar defensive behavior outside 
their burrows if prodded. 

Predators and parasites. Observations 
show that females and immature males of 
A. unicolor are preyed upon and parasi- 
tized by centipedes, pompilid wasps, acro- 
cerid flies, and mites. Adult males have 
been collected several times frorn the 
ground webs of theridiid spiders. 

Four instances of apparent centipede 
prcdation have been observed. At Clifty 
Falls, Indiana, a 50-70 mm long scolo- 
pcndromorph centipede escaped from a 
burrow containing a recently killed spider. 
At L a scolopendromorph centipede, Thea- 
tops postica, was found in a burrow with 
spider remains. At both O and Blanchard 
Springs, Arkansas, a Theatops spinicauda 
centipede was found in an empty A. uni- 
color burrow. One instance of unsuccessful 
centipede prcdation was observed at C 
where a nearly dead Lithohitis foiiicatus 



centipede (lithobiomorph) was found in 
the burrow of a large immature A. uni- 
color. The centipede had probably entered 
the burrow through the soil; the burrow 
entrance was sealed shut. 

Twenty-one cases of prcdation by pom- 
pilid wasps have been observed (Table 9). 
No larvae could be reared to adulthood, 
thus no specific or generic identification 
could be made. More than one species is 
probably involved since variation in cocoon 
size is great. Spiders with attached eggs 
or early instar larvae were found in the 
bottom of their burrows dorsal side up and 
partially ]3aralyzed. Leg jerking was the 
chief response to disturbance. The wasp 
egg was always attached centrally on the 
abdominal dorsum. Soil plugs apparently 
constructed by the ovipositing female wasp 
sealed the spider and larva into some biu-- 
rows (Fig. 15). Cocoons were usually 
attached to the burrow wall some distance 
from the burrow end, where the spider re- 
mains were usually found. The hollow 
usually found in the burrow wall just 
above the soil plug appears to be the 
source of the plug. 

One tiny, recently hatched pompilid 
larva and its paralyzed food source from 
C were placed in a narrow glass vial and 
observed through cocoon fonnation. Feed- 
ing lasted four days. The spider remains 
consisted of a small packet of compacted 
spider cuticle. On the fifth day, the larva 
had moved away from the spider remains. 
It first spun a loose maze of silk threads 
between the walls of the tube and then 
constructed the cocoon within this maze. 
Cocoon construction took less than 24 
hours. 

Population P was parasitized by the 
dipteran Eulonchu.s marialiciae Brimley of 
the Acroceridae. Observations were made 
on 1 and 2 August. This is presently the 
only host record for the genus Eulonchiis, 
which, because its distribution is other- 
wise limited to western North America 
( Schlinger, 1966 ) , and, because it is similar 
to that of the antrodiaetids, may be an im- 



282 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



portant parasite of other antrodiaetid spe- 
cies. Considering that collected specimens 
of E. morioUciae are extremely rare (C. W. 
Sabrosky, personal communication), the 
population discovered at P was surprisingly 
dense. One freshly emerged adult was col- 
lected vibrating its wings just inside an A. 
tinicolor burrow entrance. One pupa was 
collected from a burrow and kept for 
several hours until the adult emerged. 
Several other burrows were found with 
abandoned pupal skins and spider remains. 
A total of about 18 adult acrocerids were 
seen hovering near A. uniculor burrow 
aggregations. 

The adults appeared to fly only during 
the daytime. The largest aggregation of 
adults ( about 12 ) fomied rather quickly as 
I was excavating several burrows and 
hovered close to the ground where I was 
digging, as if attracted by some chemical 
released by my digging. Twice I observed 
adults hovering over and momentarily 
landing near (or on?) a closed burrow 
entrance. It is probable that, as in other 
acrocerids, the first instar larva actively 
searches out the spider, penetrates the 
cuticle, and develops as an endoparasite. 
My observations indicate that the larva, 
after feeding on the spider in the bottom 
end of the burrow, climbs up the burrow 
wall, attaches somewhere above the bot- 
tom end, and completes development in 
this spot. ( One pupa and three pupal skins 
wcr(> attached here. ) 

Immature parasitic mites have been 
found on 18 adult females from nine dif- 
ferent localities scattered over the species 
range. These were usually attached to 
appendages as well as to the body proper. 
It is likely that mite parasites on other 
specimens have gone unnoticed. 

Prey capture behavior. The normal 
foraging posture of A. unicolor is as follows 
( Fig. 59 ) : The spider is positioned inside 
the burrow just below the opened collar 
and faces out the entrance with its 
pedipalps and first legs extended slightly 
forward and touching the inner surface of 



the open collar. Usually the spider's venter 
is against the ventral-most sector of the 
burrow wall. Sometimes a spider may be 
in this posture with the collar only partly 
opened. Often prey capture behavior is 
also released in spiders which are po- 
sitioned farther down the burrow or which 
have their collars closed. Such individuals, 
however, appear to respond more slowly 
to prey stimuli. 

Larger immature spiders and adult fe- 
males usually open their collars and assume 
the typical foraging posture only in very 
reduced daylight and in darkness. Younger 
spiders, particularly the youngest, fre- 
quently leave their collars open and often 
assume the foraging posture during day- 
light hours as well. The less light (heavily 
shaded habitats, overcast days, early morn- 
ing and late afternoon hours) and the 
greater humidity, the more likely a given 
spider will be found with an open collar 
or in a foraging posture. A. unicolor is thus 
primarily a nocturnal feeder with some 
tendency to feed also during daylight 
hours, depending upon environmental con- 
ditions, age, and probably nutritional con- 
dition. 

Observations indicate that the prey 
sensing radius of A. unicolor does not ex- 
ceed several centimeters. Substrate vi- 
bration transmitted via the burrow lining 
and collar is apparently the only stimulus 
used by the spider to detennine the pres- 
ence and location of prey. This conclusion 
is indicated by numerous observations (in 
nature and in the laboratory) of prey 
capture behavior being released simply by 
the observer gently vibrating the collar 
or adjacent ground surface with forceps 
or sticks, thus eliminating visual and 
chemical cues. Additional evidence also 
rules out visual signals. In the typical 
foraging postiu-e, the spider's field of vision 
is only a limited inverted cone of space 
projecting out from the biurow entrance; 
when the great majority of prey is caught, 
it has not entered this cone. I have fre- 
quently observed spiders within closed 



Systematics and Biology of Antrodiaetvs 



Coijle 



283 



collars sensing and capturing ground sur- 
face prey. Prey capture is commonly per- 
formed in the dark; moreover, the eyes of 
antrodiactids are quite small. 

Casual observation in nature and in the 
laboratory, as well as motion pictures of 
prey capture behavior in lab animals, in- 
dicate the following pattern for prey cap- 
ture behavior. After having determined 
the approximate location of the prey by 
tactile stimuli, the spider lunges from the 
burrow entrance at the prey. Legs IV and, 
usually, III remain within the burrow 
anchoring the spider to the burrow wall. 
The pedipalps and legs I are held out in 
front of the spider, and legs II also usually 
leave the burrow. The tarsi and penulti- 
mate segments of the pedipalps and legs I 
contact the prey and hold it against the 
substrate (Fig. 60). Lateral rows of strong 
suberect macrosetae on the tarsus and tibia 
of the pedipalps and on the metatarsus 
of legs I and II aid in holding the strug- 
gling prey and probably also provide im- 
portant sensory input. Sometimes legs II 
aid in holding the prey. While the prey is 
held against the substrate, the chelicerae 
spread apart laterally, the fangs are ex- 
tended, and the spider strikes forward and 
down at the prey with the chelicerae. The 
fangs contact the prey at two well-sepa- 
rated points (Fig. 61). There may be one 
to several more such strikes and maneuver- 
ing of the prey with the pedipalps and 
legs before the prey (usually still strug- 
gling) is quickly pulled down into the bur- 
row by the pedipalps and legs I (Fig. 
62). The spider does not usually close the 
collar until later, presumably after the prey 
has been subdued. Sometimes, however, 
friction of the prey or spider legs against 
the collar as the prey is being pulled into 
the burrow partly or completely closes the 
collar. The time from the beginning of the 
lunge to the disappearance of the cap- 
tured prey into the burrow for seven filmed 
prey captures of seven different spiders 
ranged from 0.8 to 4.5 seconds, with a 
mean of 1.8 seconds. The time from the 



beginning of the initial lunge until contact 
was made with the prey for four of these 
spiders ranged from 1/32 to 1/16 of a 
second. 

Atkinson ( 1886 ) described the capture 
of an ant. After detecting the ant, the 
spider (juickly closed the collar, leaving it 
open only a crack. When the ant crossed 
this opening, the spider threw open the 
doors and caught the ant. Such behavior 
was observed several times, and Atkinson 
concluded that the spider did not strike 
until it could see the ant through the 
opening. That vision actually did play 
such an important role is doubtful. 

Prey composition. Records obtained from 
prey rejectamenta and freshly killed prey 
found in burrows in nature indicate that 
A. unicoJor is markedly unselective in its 
choice of prey. It appears to capture and 
feed upon any arthropod that it is capable 
of catching and killing (speed, strength, 
and size of the potential prey probably 
being important factors) that chances to 
enter within the prey-sensing radius of 
the burrow entrance. The prey records in 
Table 10 indicate that ants and beetles 
form a large part of this species' diet. This 
is not surprising considering the abundance 
of these arthropods on the ground surface. 
Howe\'cr, thinner-cuticled arthropods are 
probably more common prey than these 
records indicate; the rejectamenta records 
are strongly biased toward such thick exo- 
skeleton forms as ants, beetles, and milli- 
pedes which are not as finely fragmented 
by the spider. 

The ants consisted of both formicine 
and myrmicine species, \\'ith the large- 
bodied formicine genus Camponotus com- 
mon. Numerous beetle families were repre- 
sented. Three millipede families were 
found: Parajulidae (five individuals), 
Xystodesmidae (two individuals), and 
Polydesmidae (one Scytonotus gramtlatus, 
one Pseudopolydesmiis sermtiis, and two 
Psetidopolydesmiis sp. ). One of the spider 
prey was a conspecific male; another was 
a male Agelenopsis. Two phalangids be- 



284 



Bulletin Museum of Co7nparative Zoology, Vol. 141, No. 6 



longed to Leiobunum. One wasp was a 
Vespula. In the laboratory, A. nnicolor 
readily captures and eats Porcellio isopods 
and Tene])rio beetle larvae. Small im- 
niatures take wingless DrosopJiiki. Identi- 
fiable prey ranged in length from about 2 
mm to a 16-mm carabid beetle and a 25- 
mm Pseiidopohjdcsmus millipede. It is 
noteworthy that A. unicolor is capable of 
killing and digesting arthropods (ants, 
wasps, and millipedes) well known for 
their stings and/or chemical defenses. 

Disposal of prey remains. Exoskeletal 
remains of digested prey are apparently 
disposed of in two ways: 1) packed into 
the burrow wall, 2) discarded on the 
ground surface outside the burrow. In 
most burrows, prey rejectamenta is found 
mixed with a small amount of silk and 
compacted into the bottom end of the bur- 
row. All except the toughest exoskeletal 
portions (ant heads and thoraxes; beetle 
heads, thoraxes, and elytra) are usually 
fragmented. Sometimes similarily com- 
pacted rejectamenta is also present in parts 
of the burrow wall closer to the entrance. 
These trash packets possibly mark previous 
positions of the bottom end of that par- 
ticular burrow. Some burrows with large 
amounts of rejectamenta in the bottom end 
had collembola aggregated on and near the 
trash. 

Occasionally spiders kept in the labora- 
tory disposed of pieces of prey remains by 
placing them on the soil surface outside 
their burrows. At least some of these 
pieces were only partly digested and were 
decaying; this disposal behavior may be 
an adaptive response to such decay. Other 
evidence indicates that prey remains are 
disposed of outside the l)urrow: many bur- 
rows lack or have very little rejectamenta 
in their walls; a large amount of frag- 
mented insect cuticle lay on the ground 
surface in the midst of a dense concentra- 
tion of burrows at B. The periodic exca- 
vation to enlarge a burrow and the 
concomitant disposal of trash packed into 
the burrow end very likely explain both 



of these latter types of observations. In 
summary, the primary method of trash dis- 
posal could involve two steps: 1) packing 
the fragmented exoskeleton into the bottom 
end of the burrow and 2) discarding this 
compacted rejectamenta on the ground 
surface during burrow enlargement (or 
after a sufficient amount of trash collects 
in the burrow end). Such a method might 
mean less exposure to predation than 
\\'Ould the surface disposal of each prey 
exoskeleton separately. 

Mating behavior. Only a single obser- 
vation exists on the time of day of male 
wandering; I collected a wandering male 
(637 at F) on 3 August at 9 p.m. (after 
dark), 90 minutes after a heavy rain. The 
mostly digested remains of an adult male 
A. unicolor were found in the burrow of a 
conspecific female (543 at I). 

Antrodiaetus robustus (Simon) 
Ecology 

Geographic distribution and elevation 
range. Eastern Ohio east through Penn- 
sylvania, northern Virginias, and Maryland 
(Map 1). 0-1500 ft (0-460 m). 

Habitat. I have observed A. robustus in 
only one locality; Caledonia State Park in 
southern Pennsylvania. Small clusters of 
burrows were scattered over steep, well- 
shaded banks on a hillside near a stream 
in mixed conifer-hardwood forest. White 
pine and hemlock were dominant near the 
burrow sites. Most bmrows were under 
the shelter of trees, tree trunks, or exposed 
rocks and roots, and were as common in 
spots with a thick surface mat of pine 
needles as where surface litter was absent. 
The soil consisted chiefly of firmly packed 
light tan loam with pebbles. 

Life History 

Males. The data in Figure 1 indicate 
that male wandering and mating occur 
during September and October. 

Females. The second instar morphology 
of A. robustus is like that of A. unicolor. 



Systematics and Biology of Antrodiaetus • Coyle 



285 



Four records (Fig. 2) indicate that the 
timing of brood development may be 
similar to that of A. tinicolor. The presence 
of a fully sclerotized second instar brood 
within a parent burrow at Canton, Ohio, 
on 19 October weakly indicates that broods 
may overwinter within the parent burrow 
before dispersing. The sizes of the two 
completely collected broods are given in 
Table 3. 

Behavior 

Burrow structure. The five adult female 
burrows observed (Table 6) were similar 
in structure to A. tinicolor burrows. Tliey 
were widened just below the entrance and 
at the bottom end, and entirely or almost 
entirely lined with silk. The silk lining was 
thickest at the upper end. Burrow pitch 
ranged from to 10°. All burrows sloped 
downward from the entrance. The burrow 
of one immature had a dead-end side 
branch near the bottom end, another im- 
mature burrow had a similar side branch 
near the surface. 

Entrance structure. Tlie silk lining of 
the burrow is extended above the soil sur- 
face to form a flexible collar like that of 
A. unicolor. This collar is well camouflaged 
and is opened and closed in the same man- 
ner as that of A. unicolor. 

Egg sac structure and position. Both egg 
sacs observed were constructed and po- 
sitioned like A. unicolor egg sacs. Both 
were in the bottom half of the burraw on 
the ventral sector of the burrow wall 
(Table 6). 

Behavior of brooding females. Both fe- 
males with egg sacs (collected near mid- 
night) had their collars closed and sealed 
lightly on the inside with silk. 

Molting behavior. Fragmented exuviae 
were sometimes found to be bound with 
silk and compacted with prey rejectamenta 
into the bottom end of burrows. 

Defensive behavior. All individuals in 
foraging posture at night retreated down 
the burrow when disturbed by light or 



vibrations. Adult females and often im- 
matures assume an aggressive defensive 
posture like that of A. unicolor. Both 
brooding females were particularly aggres- 
sive. 

Feeding behavior. The foraging posture 
of A. robustus is like that of A. unicolor. 
Individuals assumed this stance (with the 
collar open) only in twilight and at night. 
Fragmented prey cuticle (primarily of ants 
and beetles) was found packed into the 
bottom ends of most burrows. 

Antrodiaetus pacificus (Simon) 
Ecology 

Geographic distribution and elevation 
range. The Pacific Coast of North America 
from San Francisco Bay north to southern 
Alaska, with outlying, perhaps isolated, 
populations in the montane areas of west- 
em Idaho, northeastern Oregon, and ad- 
jacent Washington (Map 2). 0-7500 ft 
(0-2280 m). The range of A. pacificus 
extends much farther north than that of 
any other North American mygalomorph 
spider. 

Habitat. The habitat requirements of 
this species appear to be similar to those 
of A. unicolor. Optimum conditions for A. 
pacificus are apparently found in humid, 
cool, rather densely forested habitats with 
deep loam or sandy loam soils. Such habi- 
tats are particularly abundant in the moist 
lowlands and mountains along the Pacific 
Coast from northern California into Alaska 
(Fig. 63), but are less continuous inland, 
there to be found in the mountains of 
Idaho, eastern Washington, and eastern 
Oregon. 

A. pacificus also occurs in the volcanic 
"sand" soil of the Cascade Mountain 
Range, and even constructs burrows in de- 
caying logs in Douglas fir forest. Burrows 
are sparse in rocky areas with shallow, fast 
drying soils. The densest aggregations 
were discovered in soil that was humid and 
firm at or just below the surface. Within 
favorable habitats, burrows tend to be 



286 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



clustered in sheltered spots protected from 
erosion and flooding. Such spots are found 
on the stable parts of slopes and banks 
along ravines, streams, trails, and narrow 
roads. Moss-covered bank surfaces, or 
portions under the shelter of root, log, rock, 
or tree overhangs are, as in A. unicolor, 
common sites of burrow aggregations. 

The densest and largest population was 
found at I where one burrow cluster con- 
tained four adult females, one adult male, 
and 36 immature spiders within 0.25 m-. 
The largest and densest population in the 
eastern part of this species' range was 
found at U. 

Life History 

Males. The data in Figure 1 indicate 
that male wandering and mating occur 
between early June and early November, 
with peak activity perhaps from late July 
to late September. Records from a few 
relatively well-collected localities hint that, 
as in A. tmicolor, male emergence is more 
nearly synchronous in any given locality 
and is controlled by local climate. 

Females. Close examination of the four 
A. pacificus broods indicate that the pat- 
tern of early postembryonic development 
is the same as in A. unicolor. Almost all of 
the shed chorions examined contained the 
embryonic cuticle, and very few of the 
spiderlings actually in the process of hatch- 
ing had lost the chorion but not the 
embryonic cuticle. Therefore hatching in- 
volves the splitting and peeling back of the 
chorion and then the embryonic cuticle in 
close succession, followed by the (usually) 
simultaneous shedding of these mem- 
branes. The morphology of the embryonic 
cuticle and the first instar is as in A. 
unicolor. 

The few data (Fig. 2) indicate that the 
timing of brood development of A. paci- 
ficus in nature may be similar to that of 
A. unicolor. Evidence that spiderlings may 
overwinter in the second instar and that 
adult females are long-lived was provided 



by an adult female (2246 at R) collected 
on 3 July. Although she was rearing no 
brood that year, her burrow was in the 
center of a 30 cm diameter aggregation of 
36 burrows of spiderlings that matched the 
size and moq:)hology of second instar A. 
unicolor spiderlings. Since she was isolated 
from other adult burrows, these spiderlings 
almost certainly belonged to her brood of 
the previous year. Data on brood size and 
egg size are given in Tables 3 and 4. 

Behavior 

Burrow structure. A. pacificus burrows 
(Figs. 16 and 17) are very similar in struc- 
ture to A. unicolor burrows. The tubular 
burrow is slightly widened just below the 
entrance and at the bottom end and is 
narrowed between. The silk lining is thick- 
est in the upper end, is often nearly im- 
perceptible in the lower portion of the 
burrow, and tends to be thicker in less 
stable or dryer soil. There is considerable 
variation in burrow length among adult 
spiders (Table 6). These burrows tend to 
be longer in dryer soil, but many other 
factors also affect burrow length. Penulti- 
mate male burrows average smaller than 
those of adult females. Adult female bur- 
row pitch ranged from to 35° \\'ith most 
burrows between 0° and 15°. No burrow 
sloped upward from the entrance. 

Entrance structure. This species pos- 
sesses at its burrow entrance a flexible 
silken collar identical to that of A. unicolor 
(Figs. 18, 19, and 44). The only burrows 
observed with the collar closed and sealed 
on the inner surface with a small amount 
of silk (as is fairly common in A. unicolor) 
were those of three penultimate males ap- 
parently ready to undergo the final molt. 
Two burrows were found with soil plugs. 
In one burrow (2994 at Catherine Creek 
State Park, Ore.), the plug was in the 
entrance imder the closed collar, but in the 
other (2970 at V), the plug was positioned 
one-third of the way down the burrow. 

Egg sac structure and placement. All 



Systematics and Biology of Antrodiaetus • Coyle 



287 



four egg sacs examined were like those of 
A. unicolor (Fig. 16). Of the three egg 
sacs which I excavated, one ( 2825 at S ) 
was near the top of the burrow, and the 
other two (at /) were in the bottom half 
(Table 6). All were attached to the ventral 
sector of the burrow wall. 

Molting behavior. Fragmented exuviae 
were often found compacted like prey exo- 
skeletons into the bottom end of burrows. 
Only two recently molted spiders were 
observed; two mature males each had the 
final exuvia packed lightly into the bottom 
end of its burrow. The exuvia of one male 
(2628 at V) was slightly broken, and the 
nature of the entrance closure could not 
be detennined. The exuvia of the other 
(L-1 at /) was unbroken, and the collar 
was closed and lightly sealed inside with 
silk. 

Defensive behavior. The behavior of 
foraging individuals of A. pacificus when 
disturbed by strong light and substrate 
vibrations is like that of A. unicolor. Adult 
females, when prodded, often assume an 
aggressive defensive posture like that of 
A. unicolor. Five large immature spiders 
at R were observed at the bottom end of 
their burrows spinning a thin septum of 
silk across the burrow lumen between 
themselves and the entrance as I excavated 
their burrows. When I broke the septa and 
prodded them, they exhibited the typical 
aggressive defensive posture. 

Predators and parasites. Three recently 
digested males were found hanging in the 
webs of theridiid spiders at G. One male 
from Washington was taken "from the 
stomach of Biifo boreiis," a western toad. 
Three large immature specimens ( 2249 and 
2640 at R, and 2114 at M) each had large 
nematode worms filling their abdomens. 
A pompilid wasp larva was found on the 
abdominal dorsum of a female A. pacificus 
on 26 July near Friday Harbor, Washing- 
ton. 

Prey capture behavior. Adult females 
were observed in foraging position within 
their open burrow entrances only after 



nightfall, or, rarely, in very dim light; the 
burrow entrances are normally closed dur- 
ing the daytime. Immatures usually as- 
sumed foraging postures earlier in the 
evening. Very young spiders frequently 
had collars open during the daytime and 
were often in foraging postures then. The 
foraging posture of this species is similar 
to that of A. unicolor. 

Prey composition. Recently captured 
prey and identifiable prey rejectamenta 
found in the bottom of burrows include 
ants, beetles, spiders, hymenoptera (one 
vespid wasp and one Argidae), and He- 
miptera. Ant and beetle remains were far 
more abundant than those of other groups. 
A 15-mm long carabid beetle (Broscinae) 
and Canjponotus major worker ants were 
the largest prey found. 

Disposal of prey remains. Fragmented 
exoskeletons of digested prey were fre- 
quently found mixed with a small amount 
of silk and packed into the bottom end of 
burrows as in A. unicolor. Tlie thicker the 
cuticle, the less fragmented the remains. 
Live collembola were observed on such 
trash in a few burrows. 

Matinii, behavior. Males of A. pacificus 
apparently wander in search of mates only 
at night. I have observed 12 wandering 
males, seven at L and five at K, all after 
nightfall. Considerable daytime collecting 
at both localities failed to turn up wander- 
ing males. Four male specimens from W 
are accompanied by the label "on ground 
at night." Wandering males usually stop 
moving when light is shown on them. 

At L, two captured males were released 
near female burro\\\s, and each stopped 
suddenly when it came close to a burrow. 
In one encounter, the female closed her 
collar just as it was discovered by the male, 
which then used his pedipalps and first 
legs in an apparent attempt to open the 
collar. After about 30 seconds it wandered 
away. In the second encounter, a female 
lunged out of her entrance toward the 
other male, made contact with its front 
legs, and both remained motionless for a 



288 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



few seconds. Then the female retreated 
quickly into her burrow and closed the 
collar. The male moved to the entrance, 
briefly handled the collar with its pedi- 
palps and first legs, and then wandered 
away. Because of the handling of the males 
and the dim light used for observation, 
both encounters were probably abnormal. 
At /, the discovery of the first leg of a 
male A. pocifictis on the soil surface at the 
burrow entrance of adult female 2232 sug- 
gested a male-female struggle. No male 
remains were found within this burrow. 

Anfrodiaefus occulfus new species 
Ecology 

Ceo^rapliic distribution, elevation raniie, 
and habitat. This species appears to be 
limited to the relatively dry and warm 
valleys of the Willamette and Umpqua 
Rivers of western Oregon (Map 2). 0-600 
ft (0-180 m). 

Life History 

Males. Adult males of this species ap- 
parently wander and mate during late Sep- 
tember and October (Fig. 1). The data 
in Figure 3 indicate that there may be a 
partial temporal separation of the mating 
season of A. occidtus from that of sym- 
patric populations of A. pugnax and A. 
pacificiis. 

Anfrodiaefus pugnax (Chamberlin) 
Ecology 

Geograp]}ic distribution and elevation 
range. Northern Oregon, southern Wash- 
ington, and northwesteiTi Idaho (Map 2). 
0-3500 ft (0-1 100 m). 

Habitat. I have observed A. pugnax at 
only two localities. At B on 8 July, a large 
population was located in a deciduous 
woods with a dense understory of shrubs 
and herbs along a small, flowing stream. 
Sunounding the woods was nonforested 
land, chiefly farmland. Although burrows 
were scattered throughout the woods on 



both horizontal and inclined ground, dense 
aggregations were foimd only beneath the 
densest vegetation. The soil was hard- 
packed brown loam \\\\h a high organic 
content, and was dry at the surface but 
slightly humid below 10-cm depth. The 
densest aggregation was found under the 
shelter of a large shrub on horizontal 
ground covered with scattered leaf and 
twig litter. Ten adult females, five penulti- 
mate males, and about 150 immatures were 
present in a 1-m- area. The habitat at 
Lewis and Clark Trail State Park, Wash- 
ington (Fig. 64) — a dense deciduous 
riparian woods located at the bend of a 
river and surrounded by wheat fields — was 
very similar to the above habitat. The 
understory vegetation was dense, and the 
burrows were found in the more sheltered 
spots on both horizontal and sloping 
ground. The soil was firmly packed light 
brown silty loam with a slightly greater 
moisture content than at B. 

Most of the other localitv records for A. 
pugnax are in dry regions at low elevations 
near rivers or streams and therefore pre- 
sumably in riparian woods similar to those 
described above. Tlie only wooded habi- 
tats in the immediate vicinity of C are such 
riparian woods. H has similar habitats. 
However, one collection of A. pugnax at E 
was made in Douglas fir forest, a more 
humid habitat. Thus it appears that A. 
pugnax is chiefly found in the low ele- 
vation riparian deciduous woodlands of the 
dry interior portions of Washington and 
Oregon, and in the low, relativeh' dry 
forest of the vallevs between the Coast 
Range mountains and the Cascade Range 
mountains. 

Life History 

Males. Collection data (Fig. 1) indicate 
that male wandering and mating occur 
within the period from late July to Oc- 
tober. 

Females. The limited data (Fig. 2) 
indicate that the timing of brood develop- 



Systematics and Biology of Antrodiaetus 



Coyle 



289 



ment in A. piignax may be similar to that 
in A. iinicolor. All individuals from which 
the brood size (Table 3) and egg size 
(Table 4) data were obtained are from B 
and Lewis and Clark Trail State Park, 
Washington. The second instar moiphol- 
ogy is like that of A. tinicolor. 

Behavior 

Burrow structure. (All observations on 
bin'row and entrance structure and egg 
sacs were made at B and Lewis and Clark 
Trail St. Pk.) The burrow structure is 
similar to that of A. unicolor. The burrow 
is roughly tubular and slightly widened 
just belo\\' the entrance and at the bottom 
end (Fig. 20). The entire burrow is silk- 
lined, with the thickest lining in the upper 
portion. The lining is thicker in burrows 
with egg sacs, at least in the vicinity of 
the sac. Burrows vary considerably in size 
but are usually relatively small (Table 6). 
The longer burrows were in softer soil. All 
burro\\'s but one were in horizontal 
ground; all these were vertical except for a 
slight turn at the bottom end of some. The 
burrow pitch of the burrow on sloping 
ground was 10°. The five penultimate 
male burrows observed averaged consider- 
ably smaller than the female burrows, were 
completely and rather heavily lined with 
silk, and had 0-5° burrow pitches. No bur- 
rows sloped upward from the entrance. 

Entrance structure. Small immature spi- 
ders have flexible collar entrances similar 
to those of A. unicolor. The silk lining of 
the burrow is extended above the soil sur- 
face, stands erect when open, and collapses 
inward to close off and camouflage the 
entrance. Soil particles and ground litter 
fragments are incorporated into this collar. 
The burrow entrances of older spiders were 
sealed and iDlugged with soil (Fig. 20), 
making it difficult to determine the exact 
nature of the unsealed entrance. The 
closure varied from a thin septum of soil 
and silk only a few millimeters thick to a 
hard plug up to 15 mm thick, mainly of 



soil. Such a sealed entrance is very diffi- 
cult to discover visually. The inside sur- 
face of the closure is covered with a lining 
of silk continuous with the burrow lining. 

Egi^ sac structure and position. The egg 
sac structure of this species is very similar 
to that of A. tinicolor (Fig. 20). The egg 
masses varied from 7 to 9 mm maximum 
diameter and 4 to 6 mm in thickness. Six 
of the eight egg sacs were positioned on 
the ventral sector of the burrow wall, the 
other two on lateral sectors. The egg sac 
silk appears to have the same texture and 
\\'hiteness as burrow lining silk. All egg 
sacs but one were positioned just above 
the bottom end of the burrow (Table 6). 

Behavior of brooding females. Like the 
large immature and nonbrooding adult 
spiders, all females with egg sacs had 
plugged and sealed entrances. 

Predators. A female specimen collected 
at G on 15 April is accompanied by a label 
saying that she was the "prey of Dinocne- 
mis," a pompilid wasp. 

Feeding behavior. Many young im- 
mature spiders were in foraging postures 
within their open burrow entrances during 
daylight, although more exhibited such be- 
havior at night. Fragmented prey exo- 
skeletons held together \\'ith small amounts 
of silk were found packed into the bottom 
ends of several burrows. Such rejecta- 
menta from the burrow of a brooding fe- 
male at B contained parts of t\vo beetles. 

Mating behavior. A label accompanying 
a male from B reports that the male was 
wandering o\er the ground surface at 
night. 

Antrodiaetus monfanus (Chamberlin) 
Ecology 

Geographic distribution and elevation 
range. Primarily the Great Basin region 
from Utah and Nevada north to Washing- 
ton (Map 2). 2500-7200 ft (750-2200 m). 

Habitat. I have observed A. montanus 
in two localities. Near Pocatello, Idaho, 



290 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



on 26 July, a small aggregation of burrows 
^^•as located in the bank of a dry ravine in 
juniper-sagebrnsh habitat at 4500 ft (Fig. 
65). Burrows of three adult females and 
several immatures were foimd in an un- 
eroded portion of the bank imder a large 
juniper. The soil surface was covered with 
a thin layer of pieces of dead grass and 
juniper. The soil was extremely dry hard- 
packed loam down to 25 cm and dry looser 
gravel and loam below that. The bottom 
ends of the large burrows were about level 
with the ravine bed. At Crater Lake, Ore- 
gon, on 30 July, I observed a sparse popu- 
lation of A. montanii.s in open and dry 
lodgepole pine, fir, and spruce forest at 
6000 ft. The two burrows observed were 
in rather tmsheltered spots, one on hon- 
zontal ground. The soil was composed of 
\olcanic "sand," very dry and loose at the 
surface, becoming firm at 5 to 10-cm depth 
and slightly humid below 15 cm. Scattered 
burrows of A. pacificiis and Atypoides 
gertschi were also found at this locality. 

The locality records in the Salt Lake 
City, Utah, area indicate that A. montamis 
is there found in dry, non- or sparsely 
wooded habitats along the lower portions 
of canyons and in the foothills of the 
mountains. All other locality records for 
this species are likewise from relatively 
xeric habitats similar to those described 
above. 

Life History 

Males. Collection data (Fig. 1) indicate 
that male wandering and mating occur 
within the period from early August 
through early November. Chamberlin and 
Ivie ( 19.35, 1945 ) observed males wander- 
ing "in the autumn." 

Females. Very limited data for A. mon- 
tamis (Fig. 2) indicate that the timing of 
brood development may be similar to that 
of A. unicolor. The brood size and egg 
diameter of the single known brood of A. 
))i07itanus (2160 at Pocatello) are given in 
Tables 3 and 4. 



Behavior 

Burrow structure. (Based on observations 
of three adult female burrows from Poca- 
tello and one large immature female burrow 
from Crater Lake.) Burrow structure is 
((uite similar to that of A. unicolor, except 
for a considerably greater average length 
(Fig. 21, Table 6). Two of the burrows 
were completely silk lined with the thickest 
lining in the upper part of the burrow. In 
the other two burrows, the lining below the 
upper 5 cm was barely perceptible. Bur- 
row pitch ranged from 5 to 25°, and all 
burrows sloped downward. Chamberlin 
and Ivie (1933), who found a female A. 
montanus without a distinct l:)urrow and 
under a cottonwood log, later ( 1935 ) er- 
roneously stated that females of A. mon- 
tanus "ordinarily do not have burrows." 
Their observation in a later paper (1945) 
was more accurate: "The female lives in 
burrows which extend about a foot or more 
straight down into the soil. The upper part 
of the burrow is lined with silk; the bottom 
is slightly enlarged, but not bent." 

Entrance structure. (Based on observa- 
tions of two adult female burrow entrances 
at Pocatello and one large immature female 
entrance at Crater Lake.) A. montanus 
constructs a flexible silken collar like that 
of A. unicolor. 

E^ig sac structure and position. The 
structure and means of attachment to the 
burrow wall of the single egg sac observed 
(2160 at Pocatello) were like those of A. 
unicolor egg sacs. Egg mass dimensions 
were 14 X 10 X 7 mm. The sac was 
positioned on the ventral sector of the bur- 
row wall deep in the burrow (Fig. 21, 
Table 6). Chamberlin and Ivie (1945) 
also found egg sacs only in the lower por- 
tion of the burrow. 

Behavior of brooding females. When 
collected in the evening before dark, the 
single brooding female (2160 at Pocatello) 
had a closed but unsealed collar. 

Parasites. The abdomen of a large im- 
mature A. montanus (0174) from Steens 



Systematics and Biology of Antrodiaetus 



Coylc 



291 



Mountain, Oregon, was filled with a nema- 
tode worm. 

Feeding behavior. The two large im- 
mature A. montamis females from Crater 
Lake were collected after dark in foraging 
positions just inside their open collars. 
Fragmented prey exoskeleton was found 
mixed with silk and packed into the bot- 
tom ends of the burrows of all thiee adult 
females from Pocatello. Two of these trash 
packets contained the remains of 12 
medium to large-sized beetles. 

Mating behavior. Apparently A. mon- 
tamis males wander in search of mates at 
night and remain under the shelter of 
various objects during the day. One male 
from Verdi, Nevada, was collected \\'alking 
over the ground at night. Chamberlin and 
Ivie ( 1945 ) often collected adult males 
"under stones, sticks, [and] bunches of 
grass" presumably during daylight hours. 
A label with the holotype male notes that 
it was resting under a piece of sagebrush. 

Antrodiaetus hageni (Chamberlin) 
Ecology 

Geographic distribution and elevation 
range. South-central British Columbia, 
eastern Washington, and eastern Oregon 
(Map 2). 1000-4000 ft (300-1200 m). 

Habitat. I have observed A. hageni at 
only one locality — at Trail, British Co- 
lumbia, on 13 July on the slopes along a 
dry stream bed in a large nonforested ra- 
vine one mile from its junction with the 
Columbia River (Fig. 66). The vegetation 
consisted of scattered young poplar trees 
and large deciduous shrubs. Much of the 
soil surface was without plant or litter 
cover. The soil was chiefly deep sand with 
a few rocks and very little organic material. 
The upper 3-8 cm of the sand was hot, dry, 
and loose; below that it was humid, well 
packed, and much cooler. Burrows were 
found in both steeply inclined and hori- 
zontal ground, but were usually clustered 
in more sheltered spots where the soil sur- 
face was stable, such as under trees and 



shrubs. A. hageni was not found in the 
cooler, humid forest habitats near Trail. 

Near Baker, Oregon, where A. hageni has 
been collected, I unfortunately searched 
only in humid forest habitats and there 
found only A. pacificus, evidence that A. 
hageni is here found in the drier habitats 
(with sparser vegetation) at and below 
4000 feet. An adult female A. hageni from 
near Oliver, British Columbia, was col- 
lected on a dry "sandy-rocky slope" covered 
with bunch grass and sagebrush (E. Thorn, 
personal communication). All other British 
Columbia records of A. hageni are in low 
river valleys where dry habitats similar to 
this and to that along the Columbia River 
at Trail are common; it is probably safe 
to assume that the habitat at Trail is an 
approximately normal A. hageni habitat. 

Life History 

Males. Adult male wandering and mat- 
ing apparently occur within the period 
from late July through October (Fig. 1). 

Females. Two records (Fig. 2) weakly 
indicate that the timing of brood develop- 
ment is similar to that of A. iinicolor. 

Behavior 

Burrow structure. (Based on observa- 
tions of two adult female and two im- 
mature female bm'rows.) The burrow 
architecture (Fig. 22, Table 6) is similar 
to that of A. unicolor. The roughly tubular 
burrow is somewhat enlarged just below 
the entrance and at the bottom end. All 
burrows were entirely silk lined, with the 
lining of the collar and upper part of the 
burrow much thicker than the rest of the 
lining. The two penultimate male bur- 
rows observed were similar to the female 
burrows in structure. Burrow pitch ranged 
from to 35°, and all sloped downward. 

Entrance structure. (Based on observa- 
tions of one adult female and two large 
immature female burrow entrances.) A. 
hageni constructs a flexible silken collar 
very similar to that of A. unicolor (Figs. 
45-46). 



292 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



Molting behavior. An immature spider 
which molted within a glass tube in cap- 
tivity fragmented its shed exuvia several 
days after ecdysis. 

Predators. A large immature female of 
A. hagcni was collected together with a 
pompilid wasp, Friocnemis oregona Banks 
(identification, H. E. Evans), on 7 April 
at Baker, Oregon. Unfortunately, no other 
data was included with the specimens. It 
is probable, however, that A. hageni is a 
host of this wasp, since the three other host 
species records for this wasp are mygalo- 
morph spiders, one being Atypoides riversi 
(Wasbauer and Powell, 1962). 

Feeding behavior. All burrow entrances 
were closed during the afternoon of ob- 
servation. Prey remains were found packed 
into the bottom end of the burrow of one 
large immature female. The prey exoskele- 
tons were fragmented (the thinner the 
cuticle the smaller the fragments ) and held 
together with a small amount of silk. 
Identifiable prey consisted of 12 ants, two 
beetles, one dipteran, and two adult male 
spiders {Tegeneria domestica and Xysticus 
sp.). 

Antrodiaetus cerberus new species 
Ecology 

Geographic distribution, elevation, and 
habitat. Known only from the type locality 
in coniferous forest in northeastern Wash- 
ington. Approximately 2000-2500 ft (600- 
750 m). 

Life History 

Males. All four examined males of this 
species were collected in pitfall traps some- 
time during May, 1962 (Fig. 1). Three 
more males were collected sometime dur- 
ing April and May, 1962 ( W. Ivie, personal 
communication). It is thus likely that A. 
cerberus males wander and mate only dur- 
ing the spring. 

Females. Two records (Fig. 2) weakly 
indicate that brood development timing 
may be similar to that of A. unicolor. 



Antrodiaetus yesoensis (Uyemura) 
Ecology 

Geographic distribution and habitat. 
Known only from the northern Japanese 
island of Hokkaido, where it is presumably 
found in the conifer forest of that cool 
temperate climate (Map 3). 

Life History 

Males. The few records (Fig. 1) indi- 
cate that the period of male wandering 
and mating occurs during the summer 
months. 

Antrodiaetus lincolnianus (Worley) 
Ecology 

Geographic distriJ)ution and elevation 
range. Eastern Kansas and eastern Ne- 
braska (Map 1). 700-1200 ft (210-360 
m). 

Habitat. In 1928, Worley stated that the 
holotyjoe male and the three paratype 
males were collected "on clay banks near 
deciduous forest" at Lincoln, Nebraska. 
Later (Worley and Pickwell, 1931), he 
stated that these males were found "under 
logs in woods" apparently in their own 
burrows. H. S. Fitch (personal communi- 
cation) collected a wandering male near 
Lawrence, Kansas, about 35 m from a 
large patch of deciduous forest through 
which ran an intennittent stream with high 
banks. This limited evidence, in addition 
to the fact that no specimens of A. lincoln- 
ianus have been collected in localities 
removed from the western lobes of the 
eastern deciduous forest, indicates that this 
species is probably found in somewhat 
humid forest habitats. 

Life History 

Males. Male wandering and mating in 
A. lincolnianus apparently occur during 
late winter and early spring. The seven 
adult males indicated in Figure 1 were 
collected by four different collectors on 
seven different dates during four different 
years. Also, tsvo other males, which I have 



Systematics and Biology of Antrodiaetus • Coijlc 



293 



been unable to examine but which are 
almost certainly A. lincolniamis, were col- 
lected by yet a different collector, Scheffer 
(1906), in Manhattan, Kansas, on 5 April 
during yet another year. Such a scattering 
of individual collecting events makes it 
more likely that these dates indicate the 
actual wandering and mating period of this 
species. 

Females. The single gravid female record 
(Fig. 2) for A. lincolniantis hints that the 
timing of brood development may be 
similar to that of A. iinicolor. 

Behavior 

Burrow structure. Worley and Pickwell 
(1931) state that "this species is very 
similar in habits to B. pocificum Simon 
[=A. pacificus] of the Pacific Coast, con- 
structing burrows in the ground six to ten 
inches deep and lining them with silk." 

Anfrodiaefus stygius new species 
Ecology 

Geographic distribution and elevation 
range. Missouri and northern Arkansas 
(Map 1). 400-1500 ft (120-460 m). 

Habitat. I have observed A. stygius only 
at Bennett Springs State Park, Missouri, 
on 4 September. An aggregation of ap- 
proximately 100 adult and immature bur- 
rows in a 4-m- area was found on a 20-35° 
slope 3-5 m from the edge of a large 
spring-fed stream in the heavy shade of a 
mixed deciduous forest (sycamore,^ elm, 
and oak trees common). Tlie ground sur- 
face had a sparse to dense cover of leaf 
litter. The hard-packed dark brown loam 
soil was slightly humid at the surface and 
increasingly humid downward. A wander- 
ing adult male was collected near a stream 
near Warrensburg, Missouri, in a similar 
deciduous forest of oak, sycamore, and elm 
(Peck, 1966). 

Life History 

Males. The limited data (Fig. 1) indi- 
cate that adult males wander and mate 
during the fall. 



Females. All three broods observed had 
developed beyond the hatching stage, but 
every shed chorion contained the embry- 
onic cuticle, an indication that the pattern 
of early postembryonic development is 
similar to that in A. tinicolor. Also the 
morphology of the embryonic cuticle, the 
first instar, and the second instar is very 
similar to that of A. iinicolor. The limited 
data (Fig. 2) indicate that the timing of 
A. stygiiLS brood development in nature 
resembles that of A. iinicolor. The size 
of the only complete brood is given in 
Table 3. 

Behavior 

Burrow structure. The three brooding 
female burro\\'s obser\'cd were completely 
silk-lined and shaped much like A. iinicolor 
burrows: roughly tubular and slightly 
widened just below the entrance and at 
the bottom end (Fig. 23, Table 6). Three 
penultimate male burrows were shaped 
similarly but had thinner silk linings and 
were shorter (Table 6). All six of these 
burrows were slightly sinuous and nearly 
xertical, with bun-ow pitches of 15-30°. 

Entrance structure. All burrows dis- 
covered — even those of immatures — had 
their entrances sealed over and usually 
plugged with soil (Fig. 23), making it dif- 
ficult to determine the exact nature of an 
unsealed entrance, which appeared to con- 
sist of a flexible collar. Tlie upper 4-10 
mm of the larger burrows was packed with 
soil. The bottom of this plug was lined 
with silk continuous \\ith the buiTO\\' lin- 
ing, and the upper exposed surface closely 
matched the surrounding soil surface in 
texture and color. 

Egg sac structure and position. The 
shape and structure of the two A. stygius 
egg sacs observed appeared similar to 
those of A. Iinicolor. Both were positioned 
on the ventral sector of the wall in the 
bottom half of the burrow (Table 6). The 
burrow diameter at the region of egg sac 
attachment may have been greater than 



294 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



normal, but the sacs still produced a slight 
localized construction of the burrow lumen. 
The silk lining was thicker in the vicinity 
of the egg sac than elsewhere below the 
upper end of the burrow. Egg sac silk ap- 
pears to have the same texture and white 
color as burrow lining silk. 

Behavior of brooding, females and 
broods-. The burrow entrances of brooding 
females, like those of nearly all other in- 
dividuals, were sealed with soil plugs. The 
brood composed of first instar spiderlings 
was still enclosed within the egg sac, while 
both second instar broods were entirely 
outside the sac. In one of these latter two 
broods, the egg sac was no longer present; 
in the other it contained all the first instar 
exuviae, indicating that the spiderlings 
emerged from the egg sac after molting 
into the second instar. The second instar 
spiderlings were quite active. 

Defensive behavior. Some adult females 
assumed a defensive posture in the bottom 
of their burrows similar to that of A. 
unicolor. 

Predators. A pompilid wasp cocoon con- 
taining a larva was found in one medium- 
sized A. sttjsiiiis burrow. Small pieces of 
spider exoskeleton were attached to the 
outside of the cocoon. 

Disposal of prey remains. Fragments of 
prey cuticle were found packed into the 
bottom end of one adult female burrow. 

Antrodiaetus opachecus new species 
Ecology 

Geographic distribution and elevation 
range. Mountains of Arizona and New 
Mexico (Map 4). 6100-8500 ft (1850- 
2600 m). 

Habitat. I have collected A. apachectis 
at only one locality, near La Cueva, New 
Mexico, at 7800 ft in a somewhat open 
forest of yellow pine, spruce, and fir. An 
aggregation of six burrows was found 
under the overhang of a large rock on the 
slope of a wide ravine. Aspen and scrub 
oak were common in the ravine. The 



ground surface under the rock lacked leaf 
litter, and the soil was humid dark sandy 
loam with some pebbles. All other locality 
records for this species are either from 
transition zone forest or low Canadian zone 
forest. Because it is restricted to such high 
elevation habitats, the species population is 
presently fragmented into several geo- 
graphically isolated populations each on a 
forested montane "island" surrounded by 
unfavorable nonforested habitats at lower 
elevations. 

Life History 

Males. Collection data (Fig. 1) indicate 
that male wandering and mating occur 
from midsummer to early fall. 

BEHA^^OR 

Burrow structure. The burrows of only 
four medium- to large-sized immature spi- 
ders were observed. The burrow shape 
appeared similar to that of A. stygius, i.e., 
tubular but slightly widened just below 
the entrance and at the bottom end. The 
silk lining was very thin and possibly ab- 
sent over most of the burrow, but slightly 
thickened at the upper end. The burrows 
were slightly sinuous and nearly vertical, 
with a burro^^' pitch of 10-20°. 

Entrance structure. The burrow entrance 
of only one immatiue spider was observed 
and appeared to consist of a short thin 
flexible collar. 

Feeding behavior. A medium-sized im- 
mature A. apachccus was found at the bot- 
tom of its burrow feeding upon a dead, 
partly digested, 20-mm long lepidoptera 
larva covered with urticating setae. Tliis 
was at 11:00 a.ai. and the collar was closed. 

Anfrodiaefus rorefzi (L. Koch) 
Ecology 

Geogra))Jiic distribution. Central portion 
of the Japanese island of Honshu (Map 3). 

Habitat. According to Yaginuma (1962), 
A. roretzi is usually found in humid, often 
heavily shaded habitats. Yamamoto (1942) 



Systematics and Biology of Antrodiaetus 



Coylc 



295 



found a sizeable burrow aggregation scat- 
tered over a very steep bank composed of 
reddish soil. Bamboo was growing over 
the upper part of the bank. Komatsu 
( 1942 ) reported that this species is found 
in shaded humid habitats, often on inclines 
or banks. Uyemura ( 1936 ) found a bur- 
row on a steep slope in moss-covered soil 
at the base of a pine tree. Ohe ( 1966 ) 
found burro\\'s chiefly on inclines (40- 
90° ) of banks along a trail. Such burrows 
were frecjuently in the shelter of exposed 
tree roots near the top edge of these banks 
or were under the overhang of exposed 
rocks. All but a few burrows were located 
on the upper two- thirds of such banks. He 
foimd burrows both in heavily shaded 
moss-covered portions of banks and in less 
well-shaded spots where the surface soil 
was rather diy. Burrows are also often 
found in caves (Yaginuma, 1962; Komatsu, 
1961) near the entrances. Komatsu (1942) 
found an aggregation of over 100 burrows 
on an inclined surface in humid reddish 
soil in dim light just inside the entrance 
of one cave. 

Life History 

Males. The small amount of data in 
Figure 1 indicates that male wandering 
and mating may occur during winter and 
early spring. More data are needed. 

Females. Ohe ( 1966 ) obsei-ved spider- 
lings in parent burrows "from September 
on. 

Behavior 

Burroio structure. Yamamoto (1942) ob- 
served the burrows of two adult males and 
several adult females (Table 6). Tliese 
were tubular and silk lined. Burrows in 
soft soil free of obstacles were longer than 
those in harder soil with roots, and male 
l)urrows averaged shorter than adult fe- 
male burrows. The burrow pitch of the 
two male burrows was apparently about 
0-10°. Komatsu (1942) observed several 
N'crtical burrows in inclined ground, but 



most were roughly perpendicular to the 
soil surface. Ohe ( 1966 ) described and 
illustrated the variation observed in the in- 
clination, shape, and length of a large 
number of A. roretzi burrows (probably 
both adult and immature). Most burrows 
apparently were not strongly curved and 
were roughly perpendicular to the plane 
of the adjacent ground surface. Exceptions 
were the result of root or rock obstacles. 
Most burrows were approximately 20 cm 
long or shorter, but a few reached 30 cm. 
Entrance structure. A pair of photos of 
an A. roretzi burrow entrance has been 
published twice (Komatsu, 1937 and 1961). 
Drawings of the entrance can be found in 
Komatsu (1961), Yamamoto (1942), Yagi- 
numa (1960), and Ohe (1966). Descrip- 
tions, all very brief, are found in Yamamoto 
(1942), Komatsu (1942), Yaginuma (1962), 
and Ohe (1966). All these describe the 
structure as a "double door entrance." 
After studying the fine photos and draw- 
ing of Komatsu, it appears to me as though 
the double door nature of the entrance has 
been overemphasized and that the entrance 
structure is more accurately described as a 
collapsible collar which is reduced at two 
opposing portions, leaving two lateral flap- 
like sectors that operate roughly as sepa- 
rate doors. Such a structure is quite similar 
to that of the collars of some adult A. 
unicolor burrows. Clearly, additional care- 
ful field observations are required. Ko- 
matsu (1942) and Ohe (1966) observed 
that, in entrances on an inclined ground 
surface, the collar is reduced at its ventral 
and dorsal sectors with both lateral flaps 
"hinging" on opposing dorsovcntral axes. 
Such an orientation is common in A. uni- 
color. Tlie collar in Komatsu's photos is 
constructed of silk and soil, well camou- 
flasied exteriorlv, and silk lined on the 
inner surface, this silk lining apparently 
being continuous with the burrow lining. 
Ohe found that collars in moss-covered soil 
were constructed partly of moss and were 
difficult to discover. Although Ohe (1966) 
observed a few burrow entrance openings 



296 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



as large as 1.5-cni diameter, most were 
roughly 1 cm in diameter. 

Molting behavior. Yamamoto (1942) 
collected two adult males in their own 
burrows. In one, the shed cxuvia lay on 
the ventral sector of the burro\\' wall about 
one-third of the way down the burrow, and 
a maze of silk threads occupied much of 
the lower half of the burrow, with the male 
in the bottom end. The second male had 
spun a thin silk septum across the inside 
of the entrance, and no exuvia or silk 
thread mazes could be found in the bur- 
row. Yamamoto did not say whether either 
collar was closed or open. Possibly the 
second spider had sealed the inside of the 
collar shut. Clearly, additional observations 
are needed to clarify behavior during the 
final male molt of A. roretzi. 

Afypoides riversi O. P.-Cambridge 
Ecology 

Geographic distribution and elevation 
range. A coastal population in northern 
California west of the Central Valley as far 
south as the Monterey Peninsula, and a 
Sierran population in the Sierra Nevada 
Mountains and their foothills (Coylc, 
196S). These populations are probably 
geographically isolated. 200-1800 ft (60- 
550 m) (coastal population). 2000-8000 ft 
(600-2400 m) (Sierran population). 

Habitat. Coastal population: Rivers 
(1891; O. P.-Cambridge, 1883) found A. 
riversi burrows in both pine and deciduous 
forests, often in stream banks. Smith 
( 1908 ) found burrows "abundant along 
shaded streams and in thickets in the foot- 
hills and mountains" of the Coast Range. 
Gertsch's (1949) habitat description was 
similar. I have collected coastal A. riversi 
in three localities, and will summarize my 
observations below. 

A. riversi is found in the foothills and 
mountains of the Coast Range, chiefly in 
shaded forest habitats. These may be de- 
ciduous, pine, or mixed deciduous-pine 
forests. Interestingly, A. riversi is uncom- 



mon or absent from the famous "basin 
forest" of predominately coast redwood 
and Douglas fir; I searched hard in such 
habitats at three localities unsuccessfully. 
A. riversi burrows are common in the next 
highest vegetation zone, the tan oak-pacific 
madrone forest containing only scattered 
coast redwood trees (Fig. 67). Burrows 
are found chiefly on rather steep inclines 
of banks along trails, roads, and streams. 
The surface litter in these spots ranges 
from absent to quite thick. The soil varies 
from compact sandy loam to clay loam, 
sometimes with a high pebble content. 
These soils are usually dry at the surface 
during the summer, but below about 10- 
cm depth are at least slightly humid. The 
densest burrow aggregations are found on 
stable, noneroding banks. Three adult fe- 
males and 29 immatures were found in a 
0.30-m- area at C. 

Sierran population. I have observed A. 
riversi in three different locations above 
6500 ft in the Sierra Nevada Mountains. 
The habitat ranged from open pine-fir 
forest to dense pine-fir or California red 
fir forest. The densest and largest popu- 
lation was found in the red fir forest at G. 
Burrows are found on both steeply sloping 
stream and road banks and on gently in- 
clined ground. The surface litter varied 
from absent to very thick. The soil was 
brown sandy loam, dry and loose at the 
surface but becoming humid and packed 
at from 3 to 15 cm deep. The lower the 
rock content of the soil, the greater is the 
burrow density. 

Life History 

Males. Coastal popidations: Male wan- 
dering and mating in these populations 
may not normally begin until September, 
October, or November (Fig. 1), and may 
coincide with the onset of the fall and 
winter wet season. At D and F on 14 and 
15 August, I searched both day and night 
in areas of dense burrow concentrations for 
adult males but found none. Sierran popu- 
lations: In at least the higher populations 



Systematics and Biology of Antrodiaetus • Coyle 297 



(6500-8000 ft), male wandering and mat- 
ing commence earlier and probably (be- 
cause of harsh winter weather) terminate 
earlier (Fig. 1). 

There is some evidence to indicate that 
the palpal tarsi of immature males of A. 
riversi may be swollen during more than 
just the penultimate instar and that many 
of the immature males with s\\'ollen palpal 
tarsi found in late summer or fall may not 
mature until one year later. The shed 
exuvia of an immature male collected at H 
on 8 August had swollen palpal tarsi like 
those of the newly formed instar. The 
exuvia of a recently molted immature male 
collected at the same locality and time had 
unswollen palpal tarsi, whereas those of 
the new instar were swollen. Thirteen 
actively foraging immature males with 
swollen palpal tarsi were collected at G, 
H, and 7 from 6 to 9 August when adult 
males were wandering. At D and F on 14- 
15 August, 14 immature males with swollen 
palpal tarsi were trapped in foraging pos- 
tures in their burrow entrances at night, 
and seven were dug out of closed and 
usually (five) sealed burrows. All 21 were 
kept cool and humid in an ice chest, but 
only the latter seven molted to maturity. 

Females. Coastal populations: The pat- 
tern of early postembryonic development 
of A. riversi appears to be similar to that 
of Antrodiaetus unicolor. All of the many 
shed chorions observed in each of the six 
egg sacs with first instar spiderlings have 
the embryonic cuticle attached, indicating 
that both the chorion and embryonic 
cuticle split and peel off simultaneously or 
in close succession and are then shed to- 
gether. The morphology of the embryonic 
cuticle, the first instar, and of the second 
instar of A. riversi, is also quite similar to 
that of A. unicolor, with the obvious ex- 
ception that the anterior lateral spinnerets 
are not lost in the second instar of A. 
riversi. The data on the timing of brood 
development in A. riversi ( Fig. 2 ) indicate 
that eggs are laid in summer, that brood 
development reaches the second instar by 



mid-autumn, and that the spiderlings 
abandon the parental burrow before the 
following summer. Rivers (1891) observes 
without further comment that "there is a 
period of aestivation, but the cause is not 
yet investigated." Data on brood size and 
egg size are given in Tables 3 and 4. Sier- 
ran populations: Only a single Sierran 
brood was collected (at 7). Its size and 
egg size are given in Tables 3 and 4. 

Behavior 

Burrow structure. A. riversi has a 
roughly tubular silk-lined burrow which 
normally increases slightly in diameter 
from the top to near the bottom end where 
there is usually a decrease in diameter so 
that the last 2-4 cm are narrow (Figs. 24 
and 25). In the Sierran populations, this 
terminal constriction was less noticeable at 
H and 7 but was well developed at G. The 
white silk lining is usually quite thick in 
the turret and upper part of the burrow, 
becoming thinner in the bottom portion, 
and may be nearly imperceptible for the 
last few centimeters. Burrows ^^'ith egg 
sacs usually had thicker silk linings than 
those without. Burrow dimensions, which 
show considerable variation, are given in 
Table 6. As might be expected, larger 
individuals within a population have larger 
burrows, and this correlation is shown in 
the lower mean burrow dimensions of the 
Sierran populations, which have, on the 
average, smaller mature individuals than 
the coastal populations. Also, penultimate 
male burrows average smaller than adult 
female burrows. Burrows ranged from 
vertical to near horizontal, but the burrow 
pitch ranged within 0-15°. Rivers (1891) 
also observed that the burrows "were more 
or less perpendicular" to the ground sur- 
face. 

The descriptions of coastal A. riversi bur- 
row structure by Rivers (1891), Smith 
(1908), and Gertsch (1949) are in agree- 
ment with my observations, except for the 
statement by Rivers that an adult female 
burrow "is more roomv at the base than 



298 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



at the opening." He apparently overlooked 
the terminal constricted portion. Smith, 
like myself, observed that "the tunnels are 
commonly considerably reduced in diam- 
eter for the last tvvo centimeters. . . ." 

Entrance structure. The silk lining of 
the burrow is extended above the soil sur- 
face to form a rather rigid elongate collar 
or turret (Figs. 26, 47-50), hence the com- 
mon names "Californian turret builder" 
( Rivers, 1891 ) and "the turret spider" 
(Gertsch, 1949). This turret is smooth 
white silk on its inner surface and is 
camouflaged and kept erect by attached 
soil particles and usually also organic ma- 
terials (from the ground surface litter) 
such as pieces of twigs, leaves, needles, 
bark, and moss. Normally the turret wall 
is supported at its base by the thickest 
accumulation of soil and other materials 
and becomes thinner distally. The opening 
at the top of the turret is roughly circular, 
and the rim is slightly expanded to fonn a 
lip. The entrance is closed off by collaps- 
ing or folding in only the distal end of the 
turret. Sometimes, as when a spider is 
molting, this closure is sealed on the in- 
side with a small amount of silk. 

There is considerable intrapopulation 
variation in turret height, as well as a 
marked tendency for individuals of the 
Sierran populations to have shorter turrets 
than those of the coastal populations 
(Table 6). A possible reason for this dif- 
ference (other than geographic variation 
in genetically determined turret building 
behavior) is that snow accumulation may 
annually destroy the Sierran tiu-rets, 
whereas coastal turrets are probably accu- 
mulations of several years' additions. A 
few turrets in the Sierran populations in- 
corporated needles, twigs, or leaves con- 
sistently attached by their ends and radiat- 
ing out roughly horizontally from the turret 
(Fig. 50). This radial arrangement closely 
resembles the "twig-lining" arrangement 
found in some of the Australian aganippine 
trapdoor spiders (Main, 1957a), but at 
present there is no evidence that in A. 



riuersi it has a similar functional signifi- 
cance in prey detection. 

The descriptions of coastal A. riversi 
turret structure bv O. P.-Cambridge (1883), 
Rivers (1891), Smith (1908), and Gertsch 
(1949) are in agreement with my obser- 
vations. Both Rivers and Smith describe 
considerable variation in the litter material 
used in turret construction. Smith includes 
two photographs illustrating some of this 
variation. Rivers oliserved turrets as tall 
as 3 inches (=7.6 cm). 

Egg sac structure and position. The egg 
sac structure ( Fig. 24 ) and egg mass shape 
of A. riversi is very similar to that of Antro- 
diaetus unicolor. Mean dimensions for the 
three egg masses are 10 X 9 X 6 mm. Egg 
sac silk appears to have the same texture 
and whiteness as burrow lining silk. In 
each of the nine burrows with egg sacs, 
the sac was positioned about half way 
down the burrow (Table 6). In five bur- 
rows the egg sac was on the ventral sector 
of the burrow wall, in three on the dorsal 
sector, and in one on a lateral sector. Some 
burrows may have been widened some- 
what in the region of egg sac placement, 
but each sac produced a localized con- 
striction of the burrow lumen. 

Behavior of brooding females and 
]) roods. All of the first instar broods were 
completely enclosed within egg sacs, but 
the thickness of the silk of three of these 
egg sacs had been much reduced and the 
spiderlings could be seen moving around 
very slowly within. It would be interesting 
to know whether these spiderlings (or the 
parent) possibly possess silk-digesting en- 
zymes. All nine l)urrows with egg sacs had 
open turrets. Three of these were found 
at night, each with the female in foraging 
position in the top of the turret. It there- 
fore appears that brooding females remain 
active at least until the spiderlings emerge 
from the egg sac. 

Molting behavior. Molting behavior in 
A. riversi appears similar to that of 
A. unicolor. Four recently molted im- 
mature spiders were collected during the 



Systematics and Biology of Antrodiaetus 



Coyle 



299 



second week of August. In each case the 
top of the turret was closed and the edges 
held together by a thin layer of silk spun 
over the inside of the closure. The recently 
shed exu\'ia was attached to the \'entral 
sector of the burrow wall two-thirds to 
three-fourths of the way down the burrow, 
venter against the wall, anterior end head- 
ing up the burrow, and with legs and pedi- 
palps extended on the wall. Tlie unsclero- 
tized spiders were in the very loottom end 
of the burrows. Five penultimate males 
(from D and F) that later molted in cap- 
tivity and several other immature males 
with swollen palpal tarsi were collected at 
other localities in similarly closed and 
sealed burrows. A fully sclerotized adult 
male ready to emerge was collected at 7 
in a sealed burrow with the last exuvia 
packed into the bottom end of the burrow. 
Defensive beJmvior. Wlien spiders in 
foraging positions were disturbed by light 
or substrate vibrations, they retreated 
rapidly down the burrow without closing 
the top of the tun'ct. During the later 
stages of excavating to expose a burrow, 
one finds the spider backed up tightly into 
the narrowed bottom end of the burrow 
in a stereotyped defensive posture. The 
cephalothorax is tilted backwards at ap- 
proximately a 45° angle with the abdomen 
and is therefore in contact with the sector 
of the burrow wall dorsal to the spider. 
Legs IV and III, and possibly II, hold onto 
the burro\\' wall to anchor the spider, and 
the other legs and pedipalps are spread 
apart laterally. The chelicerae are also 
elevated and spread apart laterally with 
fangs extended. If prodded, the spider 
strikes with a sudden downward movement 
of the cephalothorax and chelicerae. Penul- 
timate males also exhibit this same be- 
havior. Such a behavior pattern, like that 
of A. tinicolor, obviously confines the ap- 
proach of an attacker to that part of the 
spider protected by its chelicerae, but the 
narrowed burrow end of A. riversi prob- 
ably provides more effective protection for 
its abdomen than the enlarged burrow 



ending of A. tinicolor. Smith (1908) ob- 
served that the terminal constriction makes 
"a snug fit for the spider, and here it usu- 
ally snuggles down tightly when the 
digging [of the collector] has reached the 
limit necessary to secure the specimen." 

Predators. Wasbauer and Powell ( 1962 ) 
observed predation on A. riversi at Felton 
(Santa Cruz Co.), California, on 5 May, 
by the pompilid wasp Priocnemis oregona 
Banks. (See Antrodiaetus hageni predator 
records. ) 

Feeding behavior. The foraging posture 
of A. riversi appears similar to that of A. 
iinicolor; the spider is within and facing 
up the turret with the pedipalps and legs I 
extending fonvard, touching the silk lining 
on or below the turret lip. No adult spider 
was ever found in this foraging position 
until the arrival of full darkness. Unlike 
other species of Atypoides and Antro- 
diaetus, most individuals of A. riversi were 
found to leave their turrets open both day 
and night. 

A. riversi, like A. tinicolor, apparently 
detects prey primarily, if not solely, 
through the substrate vibrations trans- 
mitted via the silk lining of the tiuret and 
burrow. I was able to elicit prey capture 
responses by gently vibrating with forceps 
the ends of turret litter. The needles, 
twigs, leaves, and other somewhat linear 
litter that are often incorporated into the 
turret of A. riversi probably extend the 
prey-sensing radius of the spider even 
when, as is usually the case, this litter is 
not arranged radially. Ho\\'ever, it remains 
to be shown whether or not attaching 
linear litter to the turret is a result of 
selection for improved prey sensing. 

Fragmented prey cuticle is often found 
packed into the bottom end of the burrow 
and sometimes in other parts of the burrow 
wall as well. A cursory examination of 
trash indicates that ants may make up a 
large part of the prey of this species. 
Identifiable portions of several Campo- 
notiis workers were found. 

Smith (1908) briefly commented on the 



300 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



feeding behavior of young immatures of 
A. riversi collected from coastal popu- 
lations and kept in the laboratory. These 
accepted both small ants and aphids, and 
always maintained a hold on the lip of 
the turret with the claws of legs IV during 
the lunge for prey. 

Courtship and matinii behavior. There 
are no published observations of courtship 
or mating in the family Antrodiaetidae. I 
was fortunate enough to witness a portion 
of mating behavior between a pair of A. 
riversi at G on 8 August 1967, at 9:45 p.m., 
about 45 minutes after nightfall. When 
first discovered, the male was positioned 
within and just below the rim of the fe- 
male's turret facing down into the burrow 
with the tarsi of legs IV on the turret rim 
(Fig. 27). The female was immediately 
below the male and facing up the burrow 
in a posture very similar to the normal de- 
fensive posture, except that her fangs were 
unextended. The male's cephalothorax was 
raised slightly and its cheliceral apophyses 
were positioned as a unit between the out- 
spread chelicerae of the female (Fig. 28). 
It appeared as though the male was apply- 
ing substantial pressure with his chelicerae, 
but the female did not struggle. I could not 
observe exactly how the pedipalps and legs 
I, II, and III of the male were iDOsitioned, 
but the pedipalps appeared to be extended 
beneath the female and legs III and prob- 
ably legs II were holding onto the burrow 
wall below legs IV. After two minutes in 
this position the female stiiiggled and was 
forced down the burrow by the male, 
which appeared to maintain his hold with 
his cheliceral apophyses. I am not certain 
that the pair was actually copulating when 
discovered. 

Adult males apparently wander in search 
of mates only at night. Even though I 
searched during both daylight and evening 
hours, the Sierran males \\'ere found wan- 
dering on the ground only after 9:30 p.m., 
at least half an hour after nightfall, and 
the largest numbers were found between 



10:30 P.M. and midnight, when the search- 
ing ended. 

Afypoides gerfschi Coyle 
Ecology 

Geographic distribution and elevation 
range. Cascade Mountain Range and foot- 
hills from southern Oregon south and east 
into the northern end of the Sierra Nevada 
Mountains of California (Coyle, 1968). 
2000-7500 ft (600-2300 m). 

Habitat. The densest and largest popu- 
lations of A. gertschi were found above 
6000 ft in the open California red fir forest 
at N and P (Fig. 68). In this habitat, 
ground cover vegetation is nearly absent 
and the surface litter generally sparse. The 
soil is volcanic "sand" or sandy loam, dry 
and loose at the surface, but becoming 
humid and packed 5-15 cm below. Bur- 
rows were found both on level ground and 
inclined surfaces. At lower elevations 
burrows are often common in open forests 
with little ground vegetation or in non- 
forested areas of manzanita chaparral in 
volcanic "sand" soil, as long as the soil is 
humid below 5-15 cm. For instance, at 
O I found a dense population in manzanita 
chaparral only along the edge of the lake. 
The habitat at M is rather artificial be- 
cause of the great variety of imported 
vegetation, but it is wooded and the soil 
is sandy. Most of the burrows there were 
along the side of a stream (W. J. Gertsch, 
personal communication ) . 

Life History 

Males. The collecting data in Figure 1 
indicate that male wandering and mating 
occur during the summer. 

Females. Only a single brood of A. 
gertscki (from N) has been collected (Fig. 
2). The absence of broods with the many 
other females that I excavated is puzzling, 
but it is possible that brooding females of 
this species tend to keep their entrances 
closed noctumally and were missed be- 
cause of the difficulty of spotting closed 



Systematics and Biology of Antrodiaetus 



Coyle 



301 



entrances. Many females — all collected in 
late summer — contained small developing 
eggs that appeared to be timed for spring 
or summer oviposition. The morphology 
of the first instar is very similar to that of 
A. riversi. All of the shed chorions ex- 
amined contained the embryonic cuticle, 
an indication that hatching is also similar 
to that in A. riversi. The size of the known 
brood is shown in Table 3. 

Behavior 

Burrow structure. (See Table 6 for bur- 
row measurements.) The burrow is roughly 
tubular, with a slight expansion in diameter 
just below the entrance and a larger 
increase in diameter at the bottom end 
(Figs. 29-30). It is well lined with silk 
just below the entrance in the usually dry, 
loose surface soil, but the thickness of the 
lining decreases with depth as the soil 
increases in humidity and stability so that 
there is usually only a very thin lining over 
much of the rest of the burrow. The long 
axis of the upper portion of all burrows 
was nearly perpendicular to the plane of 
the adjacent ground surface. Almost every 
burrow curved to nearly horizontal at the 
bottom end. 

Entrance structure. The tubular silk 
lining of the burrow extends above the 
ground surface to fomi a rather thick but 
flexible collar which is collapsed inward 
to close off the entrance (Figs. 31-34; 51- 
53). The collar is composed chiefly of soil 
particles and small bits of surface litter 
held together by silk so that a closed en- 
trance is remarkably well camouflaged. 
When open, this collar normally does not 
remain erect, but collapses outwardly so 
that it is roughly parallel to the ground 
surface. Usually only the proximal portion 
of the inner surface of the collar is a clean 
white extension of the burrow lining. The 
burrow entrance opening is often slightly 
elliptical. As in Antrodiaetus iinicoJor, a 
closed collar usually gives the appearance 
of a "double door" (which it is not) be- 



cause the spider collapses the collar in a 
bilateral manner by pulling inwardly on 
two opposing (lateral) sectors, producing 
a condensation of folding at the two op- 
posing points where these sectors meet. 
The collar is often somewhat reduced at 
these two opposing points, which cor- 
respond to the ends of the long axis of an 
elliptical entrance opening. No sealed col- 
lars were found. 

Egg sac structure and position. The only 
A. gertschi egg sac collected appeared to 
be very similar in size, shape, and con- 
struction to those of A. riversi (Fig. 29). 
It was attached to the ventral sector of the 
burrow wall less than one-third of the way 
do\\'n the burrow at a point where the bur- 
row appeared to have been slightly en- 
larged, but the sac produced a localized 
constriction of the burrow lumen. 

Molting behavior. A recently molted fe- 
male spider (2016 at M) was collected on 
31 July. The burrow entrance was closed 
but unsealed, and a plug of soil had been 
placed one-third of the way down the bur- 
row. The weakly sclerotized spider was in 
the bottom of the burrow. The exuvia, 
heading up the burrow and with legs ex- 
tended, was lightly attached to the silk 
lining of the ventral sector of the burrow 
wall several centimeters from the bottom 
end. 

Defensive behavior. When a female in 
foraging position was disturbed by light 
or substrate vibration, she either rapidly 
withdrew down the burrow without closing 
the entrance or suddenly closed the collar 
with her pedipalps and anterior legs and 
held the collar shut without retreating. 

Feeding behavior. Burrow entrances 
were closed during daylight hours. Tlie 
youngest spiders opened their entrances 
and assmned a foraging posture earlier 
than older spiders, often as soon as it be- 
came twilight. Mature females usually 
opened their entrances well after nightfall. 
The foraging posture of A. gertschi ap- 
pears similar to that of A. riversi, with 
pedipalps and anterior legs resting on the 



302 Bulletin Museum of Compaiativc Zoology, Vol. 141, No. 6 



basal portion of thc> collar. If the plane of 
the burrow opening is not horizontal, the 
spider usually orients this stance with its 
venter against the lowest sector of the 
burro\\' opening, which is also usually the 
long end of an oval entrance opening. 
The exoskelctons of two freshly digested 
Camponotus worker ants were found in 
one adult female burrow, and ant pieces 
were observed on the ground surface next 
to another burrow. Only occasionally was 
prey cuticle found in burrows. 

Couriship and matinii, behavior. Males 
of A. iiertschi apparently wander only after 
dark. All wandering males (14) that I 
have collected were found after 10 p.m. 
even though collecting included both day- 
light hours and the period from 8 to 12 
P.M. A female at O was discoxered at 11 
P.M. feeding on a recently killed con- 
specific male several centimeters below 
her burrow entrance. 

Atypoides hadros Coyle 
Ecology 

Geographic di.str Unit ion and elevation 
range. Southern Illinois and eastern Mis- 
souri (Coyle, 1968). 400^1100 ft (120- 
330 m). 

Habitat. I have observed A. hadros at 
only two localities. At Feme Clyffe State 
Park, Illinois, burrow aggregations were 
found in shallow shelter caves cut into the 
base of steep limestone bluffs. Mixed hard- 
wood forest extended along the base of 
these bluffs. Burrows were found only in 
those shelter caves kept moist by water 
seepage in late summer. The largest aggre- 
gation of burrows was found on the floor 
of the most humid shelter, where males, 
females, and immatures of both A. hadros 
and Antrodiaettts imicolor lived side by 
side. The soil was a moist to wet mixture 
of soft clay-loam and pebbles. Surface 
litter was absent. Burrows were present 
both at the well-lighted edges of the shel- 
ter caves and in the dimly lit areas. Except 
in the most humid shelter cave, A. unicolor 



uas more abundant than A. hadros. At 
Montauk State Park, Missouri, burrows of 
both A. hadros and Antrodiaetus were 
found side by side in moist reddish clay- 
loam with very little surface litter at the 
base of a high rock outcrop on the bank 
of the Current Rixer in mixed hardwood 
forest. 

The other three localities where A. 
hadros has been collected (Little Grand 
Canyon, Pine Hills, and Lusk Creek, Illi- 
nois) are also in forested areas where 
ravines, rock outcrops, and other sheltered 
spots are common. An adult female from 
Little Grand Canyon was taken from a 
burrow in a drier, less sheltered spot than 
the shelter caves of Feme Clyffe State 
Park (J. Beatty, personal communication). 
A. unicolor is also apparently more abun- 
dant at each of these three localities than 
is A. hadros. 

Life History 

Males. Collecting data ( Fig. 1 ) indicate 
that penultimate males molt to maturity 
during late summer or early fall and \\'an- 
der and mate during the first half of fall. 
Two mature males were collected on 13- 
14 September in their burrows shortly after 
the final ecdysis. Pitfall traps planted at 
Pine Hills by J. Nelson in 1967 collected 
one wandering male during the first week 
the traps were set out (7-14 October), nine 
males during the next week, none during 
the next week, three during the next week 
(27 October-3 November), and none dur- 
ing the next week, the final \\'eek traps 
were put out. 

Females. The limited data (Fig. 2) hint 
that timing of brood development in A. 
hadros may be similar to that in A. uni- 
color. Second instar morphology is like 
that of A. riversi. The size of only one 
brood could be determined (Table 3). 

Behavior 

Btirroto structure. The burrows were 
roughly tubular and completely lined with 



Systematics and Biology of Antrodiaetus 



Coyle 



303 



silk. The lining of several adult female and 
both adult male burrows were espe- 
cially thick and white. Most burrows were 
widened slightly just below the entrance 
and at the bottom end (Fig. 35). Burrows 
of adult males averaged smaller than those 
of adult females (Table 6). Probably be- 
cause of the pebbly soil, many burrows 
had abrui^t curves, with the slope of die 
bottom end varying from vertical to hori- 
zontal. 

Entrance structure. The entrances of 
only two adult females and two immatures 
were observed. The tubular silk lining of 
the burrow was extended slightly above 
the ground surface to form a very short 
and thin collar (Figs. 36-37). Soil particles 
were incorporated into the collars so that 
the external surface had the same texture 
and color as the surrounding soil surface. 
The height of each collar ^^'as quite 
strongly reduced at two opposing points 
on its circumference. When the collar is 
collapsed, the folding is concentrated at 
these two points, and the opposing higher 
sectors or flaps of the collar meet and 
overlap in the center along the line con- 
necting the two reduced points. 

Brood beliavior. No egg sac remnants 
were present in the four burrows with 
second instar broods collected from Feme 
Clyffe State Park. These spiderlings re- 
acted quickly to stimuli and could move 
about rapidly. They were clustered in the 
bottom of two burrows, and were more 
scattered in the other two but retreated 
do\\'n to the burrow bottom as I exposed 
each burrow. 

Molting behavior. The burrows of both 
recently matured males were heavily lined 
with silk. Both spiders were sufficiently 
sclerotized to be active. The recently shed 
exuviae had been pushed into the very bot- 
tom ends of the burrows and covered over 
with a thin layer of silk. Since both bur- 
rows were discovered by scraping the soil 
surface, I could not determine whether the 
closed collars had been sealed with silk. 

FeecUns, behavior. Burrow entrances 



were closed during the daytime. Burrows 
of several immature A. haclros observed 90 
minutes after nightfall had the collars open 
only a crack, while adjacent A. unicolor 
immatures had their collars wide open. 
Fragmented prey exoskeletons were packed 
into the bottom ends of a few burrows. 
Such trash from two immature biurows 
was examined and found to consist of ant 
and beetle parts. 

Aliatypus californicus (Banks) 
Ecology 

Geograplnc distribution and elevation 
range. West central California from the 
San Francisco Bav region to the Monterey 
Peninsula. 1000^2900 ft (300-880 m). 

Habitat. I have collected only three 
specimens, all these near Felton, California, 
between 1000 and 1600 ft in or near aggre- 
gations of Atypoidcs riversi. Actinoxia 
burrows were also present but as uncom- 
mon as those of A. californicus. All bur- 
rows were found on steeply inclined road 
banks in rather dense mixed deciduous- 
coniferous forest (tan oak, pine, and some 
coast redwoods ) . The soil varied from clay 
to sandy loam, was quite rocky, and was 
dry at the surface. 

The holotype specimen was collected at 
the side of a stream (Banks, 1896). Smith 
(1908) collected A. californicus in the foot- 
hills and mountains on both sides of the 
Santa Clara Valley, finding burrows most 
commonly along banks of streams and roads 
in "fine compact sandy soil." Occasion- 
ally bun-ows were found in "sand\' adobe" 
soil. These banks with burrow aggregations 
"usually ha\e little or no vegetation upon 
them other than short scattered moss." 
Smith found A. californicus "commonly 
associated with Euti/chides [= Actinoxia] 
and Atypoides [riversi]." 

Life History 

Males. Smith (1908) collected the only 
two known males of A. californicus during 
October (Fig. 1), but failed to indicate 



304 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



\\ hcther tliey were taken from burrows or 
collected wandering. 

Females. Smith ( 1908 ) observed that 
most burrows (at least of those found in 
exposed banks) were sealed shut from 
June to "the first rains in December." 
"None were found with sealed doors from 
the end of December to the end of April." 
At least some burrows in shaded stream 
banks were not sealed for such a long 
time. Such estivating beha\'ior during the 
dry season may be common to other spe- 
cies of Aliatypus. 

Behavior 

Burrow structure. Smith (190<S) states 
that "the burrow is comparatively long, 
[and] simple. . . ." "I have seen no evi- 
dence of branches of any type, nor any 
suggestions of extensions above the sur- 
face. The silk lining is so meagre as to be 
practically indiscernible. The burrows of 
small specimens are not easy to distinguish 
from those of certain mining bees, tiger 
beetles, etc. . . ." Gertsch (1949) notes that 
"the silken lining is quite thin, but thick- 
ens around the opening. ..." I have ob- 
served only two burrows. One was an 
adult female burrow which was well lined 
with silk in the upper half but with very 
little or none in the lower half. The bur- 
row was nearly straight, sloped downward 
with a burrow pitch of 15°, and the bottom 
end was somewhat enlarged. The other 
(immature) burrow was straight, sloped 
downward with a burrow pitch of 30°, 
was roughly of equal diameter its entire 
length, and was detectably lined with silk 
only near the entrance. (See Table 6 for 
dimensions.) 

Entrance structure. Banks (1896) re- 
ported that the holotype burrow entrance 
was covered with a trapdoor. Smith ( 1908 ) 
observed that the entrance structure was a 
"simple trapdoor" constructed of silk and 
soil, sometimes with bits of moss or grass 
attached. Spiders inactive during the dry 
season had the trapdoors "plastered down 



firmly with soil." "Doors sealed up in this 
way are harder to see, as a rule, than when 
normally fastened at the hinge only." Smith 
gives the dimensions of two of the largest 
trapdoors he examined: 21 X 15 mm and 
18 X 13 mm. Gertsch (1949) described the 
trapdoor as being "of the wafer type." I 
have observed the entrance of only one 
medium size immature burrow. The trap- 
door was thin and waferlike but quite 
tough, and was joined broadly to the bur- 
row lining by a wide hinge. Tlie outside 
surface of the door matched the immediate 
ground surface, and the inside surface was 
silk continuous at the hinge with the silk 
lining of the burrow. Tlie trapdoor was 
closed and the entrance sealed by a 1 cm 
thick hardened plug of soil just inside the 
door. 

Trash disposal. Both burrows which I 
observed had fragmented prey rejecta- 
menta packed into their bottom ends. 

Aliatypus no. 1 
Ecology 

Geoiiraphic distribution and elevation 
ran^e. A small area in southern California 
between Riverside and Palm Springs. 
1000-4000 ft (300-1200 m). 

Habitat. A population of A. no. 1 was 
observed in a xeric habitat in the foothills 
of the Box Springs Mountains on the edge 
of Riverside, California. Burrows were 
scattered over the 30-80° inclines of ravine 
banks covered only with dry scattered 
vegetation of grass and shrubs. The domi- 
nant larger plants were wild buckwheat 
(Eriogonum) and sagebrush (Artemesia). 
The soil was chiefly orange sandy loam 
with a rather high content of decomposed 
granite. Two other genera of burrowing 
mygalomorph spiders (Bothriocyrtum and 
Aptostichus) were found in this same lo- 
cation and habitat. 

Life History 

Males. The data (Fig. 1) indicate that 
males of this species may wander and mate 



Systematics and Biology of Antrodiaetus 



Coijlc 



305 



with the coming of humid weather in late 
fall or winter. 

Females. Examination of the contents of 
four egg sacs containing first or second 
instar spiderlings indicates that the pattern 
of early postembryonic development in this 
species is similar to that in Antrodiaetus 
tmicolor. The moq^hology of the embryonic 
cuticle, first instar, and second instar is also 
similar to that of A. unicolor, with the 
exception of those differences in the 
second instar that are present in older 
instars as well. The four brood records 
(Fig. 2) indicate that the cycle of brood 
development in this species may begin 
later than that of Antrodiaetus unicolor 
and other species of Antrodiaetus and 
Atypoides. The two first instar broods 
collected on 5 October and one of the first 
instar broods collected on 10 October were 
kept alive and had reached the second 
instar stage by the end of October. Brood 
sizes are given in Table 3. 

Behavior 

Burrow structure. The burrows of four 
adult females, several immatures, and one 
adult male were obsei'ved. They were 
tubular and relatively long (Table 6) and 
straight. Variation in diameter along their 
length was not recorded. The adult female 
burrows had a thick silk lining near the 
entrance and usually a thin silk lining over 
most of the rest of the burrow. The walls 
in the vicinity of the egg sac were more 
thickly lined with silk. Immature burrows 
were silk lined near the entrance but very 
thinly or not at all elsewhere. The burrows 
of two adult females sloped downward 
with pitches of 0-10°. The male burrow 
was considerably smaller than the adult 
female burrows. 

Entrance structure. The burrow entrance 
is provided with a wafer type trapdoor 
(Figs. 38-39 and 54-55) composed of soil 
and silk and often particles of ground sur- 
face materials. The trapdoor of adult fe- 
males is 1-2 mm thick. The outer surface 



is well camouflaged and the inner surface 
is silk lined, this lining being continuous 
with the burrow lining at the hinge. The 
hinge is almost as wide as the maximum 
diameter of the entrance. The rim of the 
burrow opening is usually flared outward 
slightly to form a narrow lip against which 
the trapdoor fits snugly when closed. Many 
of the spiders collected in September and 
early October had sealed their trapdoors 
shut with silk seals and/or soil plugs ap- 
plied to the inside of the door. The adult 
male's trapdoor was sealed shut with silk 
applied to the inner surface of the door 
and the adjacent burrow wall. 

Efifi, sac structure and position. The 
structure of the egg sac is poorly known. 
The thickness of the sac wall appears simi- 
lar to that of Antrodiaetus and Atypoides 
egg sacs. The egg sac is apparently firmly 
and broadly attached to the burrow wall, 
and at least somewhat flattened, as in 
Antrodiaetus and Atypoides species. The 
sacs were positioned in the bottom of the 
burrow (Table 6). 

Moltinfi behavior. The penultimate 
exuvia of the nearly fully sclerotizcd adult 
male lay slightly broken up in the bottom 
end of the burrow. Its trapdoor was sealed 
shut with silk. 

Feedinp. Ijehavior. Broken prey cuticle 
was found in the bottom end of several 
burrows. Identifiable prey included several 
tencbrionid beetles, ants, and a mutillid 
wasp. 

Aliatypus no. 2 
Ecology 

Geographic distribution and elevation 
ran<ie. Southern Califomia in the vicinity 
of Los Angeles. 500-2500 ft (150-760 m). 

Habitat. A dense burrow aggregation of 
this species was found at Placerita State 
Park, Califomia, on a north-facing road 
bank. Burrows of three adult males, five 
brooding females, and another adult fe- 
male were found within a 0.2-m- area. 



306 



Bulletin Museum of Cornparative Zoology, Vol. 141, No. 6 



Life History 

Males. Collecting records (Fig. 1) in- 
dicate that adnlt males of this species wan- 
der and mate dnring the wet season in late 
fall and in winter. 

Females. An examination of the single 
first instar brood and the second instar 
broods indicates that the pattern of early 
postembryonic development is similar to 
that of Antrodiaetus unicolor. Almost all 
of the shed chorions examined contained 
the embryonic cuticle. Tlie morphology of 
these stages is similar to that of A. unicolor, 
except for the larger rastellar macrosetae 
and retention of the anterior lateral spin- 
nerets in A. no. 2. Brood records (Fig. 2), 
like those of A. no. 1, indicate that the 
cycle of brood development may begin 
later than in Antrodiaetus tinicolor. Only 
one complete brood was collected (Table 
3). 

Behavior 

Egg sac structure. The single preserved 
egg sac had a similar shape and the sac 
wall thickness appeared similar to that of 
Antrodiaetus unicolor. 

Behavior of brooding females and 
broods. The spiderlings apparently emerge 
from the egg sac early in the second instar 
or possibly very late in the first. Only the 
first instar brood was still within the egg 
sac. 

Aliatypus spp. 

Several other Aliatypus populations have 
been sampled, but few natural history 
observations have been recorded. A few of 
these populations probably belong to the 
above pair of undescribed species; others 
represent new species. These samples are 
scattered widely in Southern California 
from Yosemite National Park south to the 
Los Angeles area. Their total elevation 
range is 1000-7000 ft (300-2130 m). Al- 
most all samples were collected in dry non- 
forested habitats similar to those of A. no. 
1 and A. no. 2. One population (possibly 



conspecific with A. no. 2 ) was found along 
the 30-60° incline of a 2-meter deep ravine 
in Water Canyon in the Tehachapi Moun- 
tains, California. Grasses and willow 
bushes (Salix) were the dominant vegeta- 
tion. The soil was dark friable clay-loam 
that became humid at about 20 cm below 
the surface (on 7 September). Burrows of 
the ctenizid genus Aptostichus were pres- 
ent in the same ravine. All male records 
( Fig. 1 ) indicate that wandering and mat- 
ing occur during the winter wet season. 
When observed, entrances have always 
been trapdoors. 

Summary and Discussion 

So many gaps exist in our present knowl- 
edge of these spiders that it is not possible 
to do much more than briefly summarize 
and discuss the above data where it is 
plentiful or of special interest. 

Ecology 

Habitat. Throughout the major part of 
their ranges, the genera Antrodiaetus and 
Atypoides are clearly the most successful 
(in terms of abundance) mygalomoiph 
spider genera. Apparently few other 
mygalomorph groups are so successfully 
adapted for life in the cool temperate zone. 
The different species of Antrodiaetus oc- 
cupy a variety of habitats — at one end of 
the spectrum A. unicolor and A. pacificus 
are foimd in humid forest habitats and at 
the other end A. montanus lives in drv, 
often nonforested, habitats — but the genus 
is basically adapted to fairly humid forest 
or woodland habitats. In the northwestern 
United States, where a number of species 
are sympatric, there is evidence for habitat 
segregation among the closely related spe- 
cies A. pacificus, A. hageni, A. montanus, 
and A. pugnax, i.e., ecological isolating 
mechanisms may be operative. 

Tliere is consideral:)le evidence (de- 
scribed above) that A. hageni is found in 
relatively dry unforested habitats that are 
often distinct from the more humid forest 



Systematics and Biology of Antrodiaetus 



Coijle 



307 



habitats of sympatric A. pacificiis popula- 
tions. These two species have not yet been 
collected together in mixed burrow aggre- 
gations. A. monfanus occurs in unforcsted 
or thinly forested habitats usually too dry 
for A. pacifictis. It apparently occurs in 
mixed populations with A. pacificus only 
in open montane forests such as at Crater 
Lake, Oregon. (This is apparently marginal 
habitat for both species.) In the eastern 
portions of Washington and Oregon, the 
low, riparian, deciduous woodland habi- 
tats of A. ptignax are usually separated by 
wide expanses of wheatland or sagebrush- 
grassland from the higher montane forest 
habitat of A. pacificus. Except for possible 
contact of these two habitats along streams 
in the foothills of mountains, these two 
species may be effectively separated in 
this region. However, in the Willamette 
River Valley and the lowlands to the north, 
both species have been collected at the 
same localities. (Adult males of both spe- 
cies have been collected in Douglas fir 
forest at Olympia, Washington, and in a 
single pitfall collection at Eugene, Ore- 
gon.) Apparently habitat segregation in 
this region is at least imperfect. I have 
yet to find mixed species populations of 
any pairing of A. hageni, A. montanus, and 
A. piignax. 

The species of AUjpoides are, like most 
Antrodiaetus species, usually found in rel- 
atively humid forest or woodland habitats. 
Even though Atijpoides geiischi is some- 
times found in manzanita chaparral, it is 
foimd only where the subsurface soil is 
humid. All three species are sympatric 
with Antrodiaetus species. Atijpoides had- 
ros is commonly found in mixed species 
populations with Antrodiaetus tinicolor, 
with no evidence yet of habitat segrega- 
tion. Over much of its range A. gertschi is 
sympatric with Antrodiaetus pacificus and 
probably Antrodiaetus montanus. One or 
both of these species was often found in 
low numbers in A. gertschi aggregations. 
On the slopes of Mt. Shasta, California, 
above 5500 ft, only a single Antrodiaetus 



specimen was found among the dense A. 
gertschi populations, but, at 4950 ft at 
McBride Springs, A. gertschi and Antro- 
diaetus exhibited habitat segregation. 
Antrodiaetus (immature, probably A. paci- 
ficus) burrows were found only in the 
dark, humid, organically rich loam soil of 
the stream bank immediately below the 
spring in the heavy shade of a small dense 
stand of young fir and pine trees. A. 
gertschi burrows were present only in the 
volcanic "sand" soil (dry and loose at the 
surface, humid and packed below 5-7 cm 
depth) along the edge of the same stream 
in an extensive area of chaparral only 10- 
20 m downstream of the spring. 

The habitats of Aliatypus species are 
poorly known, but are relatively dry and 
usually nonforested. Most species are 
found within the Califomian Province as 
defined by botanists (Gleason and Cron- 
quist, 1964) and experience the long an- 
nual dry season characteristic of southern 
California. Aliatypus species are found 
competing in the same habitats with 
several genera of Ctenizidae, the trapdoor 
spider family towards whose niche Alia- 
typus has converged. 

Life History 

Males. In most species of Antrodiaetus 
and Atypoides, the males emerge from 
their burrows, wander, and mate sometime 
during the summer and fall, primarily dur- 
ing the second half of summer and early 
fall (Fig. 1). One exception is Antrodiaetus 
lincolnianus, which apparently mates in 
early spring. Two other possible excep- 
tions are Antrodiaetus cerberus and Antro- 
diaetus rorctzi. The partly sympatric 
species A. lincolnianus and Antrodiaetus 
stygius appear to be separated by an 
effective temporal isolating mechanism. 
Other pairs of Antrodiaetus that are pos- 
sibly separated in part by temporal isolat- 
ing mechanisms are A. pacificus and A. 
occidtus, A. pacificus and A. cerherus, and 
A. occidtus and A. pugnax. The studied 



308 



Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



species of Aliatypiis appear to mate during 
late fall or winter. 

Climate appears to have an important 
influence on the time of ^^'anclering and 
mating. High elevation populations (Antro- 
diaetus apachecus, the Sierran populations 
of Att/poides riversi, and Afypoides 
p^ertschi) often tend to mate earlier than 
low elevation relatives. There seems to be 
considerable selective pressure for male 
wandering to occur during relatively 
humid periods. Some evidence exists for 
the triggering of A. iinicolor male wander- 
ing by humid weather. Aliafiipiis species 
do not mate until the California wet 
season commences. Coastal populations of 
Att/poides riversi probably do not mate 
until the long dry season ends in the fall. 

Females. The pattern of early post- 
embryonic development appears to be 
similar in all three genera, although many 
species are unstudied. The timing of brood 
development in nature is reasonably well 
known for only one species, Antrodiaetus 
unicolor (Fig. 2). The small amount of 
data for a few other species of Antrodiaetus 
and Atypoides indicate that these have a 
somewhat similar schedule of brood de- 
velopment. In some species of Aliatypiis 
the brood cycle may be shifted slightly 
toward the beginning of the winter wet 
season, but more information is needed. 
Some species (Antrodiaetus j)Uis,nnx, Antro- 
diaetus sty<i,ius, and Aliatypiis californicus) 
may commonly seal their burrow entrances 
for rather long periods of time during the 
summer. Such inactivity may be primarily 
an adaptation to prevent desiccation dur- 
ing dry periods, or possibly to reduce 
parasitism. 

Antrodiaetus unicolor and Antrodiaetus 
paeificus have markedly larger mean brood 
sizes than any other species of Antrodiaetus 
and Atypoides studied (Table 3). Tliis 
ability to produce large broods may be an 
important factor in the success of these two 
abundant species. Aliatypiis is poorly 
studied. The egg size of A. unicolor and 
A. paeificus averages smaller than that of 



the other species studied (Table 4). It 
would seem advantageous for an individual 
spider to be able to reduce egg size 
as much as possible without reducing 
offspring survival ability and thereby 
maximize the egg number. Why have A. 
unicolor and A. paeificus, the adult females 
of which average slightly smaller than 
Antrodiaetus montanus and larger than 
Antrodiaetus pu^nax and Atypoides riversi, 
been most successful in reducing egg size? 
Perhaps a lower probability of egg and 
spiderling desiccation in their more humid 
habitats is an important factor. Much 
more and other kinds of data are needed to 
attack this problem effectively. 

Behavior 

Burrow structure. There are numerous 
factors such as substrate environment and 
body size that cause intraspecific variation 
in burrow shape and size and in the con- 
dition of the burrow lining. However, 
some aspects of burrow structure are rather 
constant within species and show inter- 
specific differences. (The three Afypoides 
species can be easily distinguished from 
each other by differences in burrow and 
entrance structure.) The burrows of all 
studied Antrodiaetus and Atypoides spe- 
cies are roughly tubular and more or less 
enlarged at the bottom end and just below 
the entrance, with the exception of 
Atypoides riversi, in which the very bottom 
end of the burrow is usually narrowed. 
Aliatypiis is poorly studied. Adult female 
burrow length varies considerably within 
species, but species of smaller body size 
(Antrodiaetus pupmx, Atypoides hadros) 
tend to have shorter bin-rows, and those in 
habitats with particularly dry soil (Antro- 
diaetus montanus, Aliatypiis no. 1) ap- 
parently tend to have longer burrows 
(Table 6). All studied species line their 
burrows with silk, with the thickest lining 
in and near the entrance, where the sub- 
strate is usually the least stable. Some 
species frequently have very little lining 
silk in the rest of the burrow. 



Systematics and Biology of Antrodiaetus • Coyle 



309 



The burrow pitches of the studied spe- 
cies in all three genera fall within 0-35° 
and average between 0° and 15°. Only 
rarely do the burrows of any studied spe- 
cies slope upward from the entrance. The 
probable adaptive significance of such a 
nearly perpendicular burrow inclination is 
that as much of the burro\\' as possible is 
removed from the unstable environment of 
the surface soil. Atypoides fi^ertschi bur- 
rows tend to turn toward the horizontal at 
their end. Burrow construction behavior 
has been observed in only one species, 
Antrodiaeius tinicolor. More complete 
studies of this behavior, particularly of 
entrance construction, are essential for 
tracing the evolution of burrovv' and en- 
trance structure in the Antrodiaetidae. 

Entrance structure. All studied species of 
Antrodiaetus have flexible collar entrances 
that stand erect when open. However, the 
collar of Antrodiaetus roretzi is apparently 
markedly reduced at tvvo opposing sectors 
so that it consists essentially of two oppos- 
ing flaps. Atypoides hadros and Atypoides 
^ertschi both have flexible collar entrances, 
but that of A. g^ertschi is rather thick, and 
when opened, collapses outward instead of 
remaining erect. Atypoides riversi con- 
structs a rigid elongate collar or turret, the 
opening of which is nonnally kept open 
both day and night. All studied species of 
Aliatypus construct wafer type trapdoors. 

Probably the two chief functions of the 
entrance closure of these spiders are to 
prevent the entry of predators and^ para- 
sites (by providing camouflage and struc- 
tural resistance) and to help maintain a 
favorable environment within the burrow 
(by preventing rain and runoff from enter- 
ing and by maintaining higher humidity 
and moderate temperature ) . Without more 
behavioral and ecological data, it is diffi- 
cult to say what special functional ad- 
vantages each type of closure mentioned 
above has. The trapdoor closes automati- 
cally when the spider retreats from the 
entrance, and probably a trapdoor more 
effectively seals out external climate and 



flooding than a collar. Perhaps the Alia- 
typus trapdoor originated as a key adap- 
tation for relatively unsheltered habitats 
in climates such as that of southern Cali- 
fornia, where occasional torrential rainfall 
with runoff and flooding and where long 
dry summers are characteristic. The pos- 
sible functional significance of the turret 
of Atypoides riversi is particularly puz- 
zling. Possibly it involves an enlargement 
of the prey sensing surface area. Possibly 
its rigidity and height serve to protect the 
burrow during runoff in heavy rain. A 
better understanding of the special func- 
tions of these closure types is needed for 
a clearer understanding of their evolution- 
ary history. 

I strongly suspect that both the turret 
entrance of Atypoides riversi and the 
double flap collar of Antrodiaetus roretzi 
have been derived from collar entrances 
resembling those of the other species of 
these two genera. It is possible that the 
trapdoor entrance of Aliatypus has been 
derived from a collar via the development 
of a dominant flap. It presently seems just 
as likely that both this trapdoor and the 
collar were derived independently from a 
simpler, more generalized ancestral type 
of entrance. It seems much less likely that 
the collar has been derived from a trap- 
door entrance. 

There appear to be two methods (per- 
haps not always distinct) used for sealing 
an entrance: 1) sealing inside with silk 
only, or 2) sealing inside with a plug of 
soil and usually silk combined. Both 
methods have been observed in each of 
several species (Antrodiaetus unicolor, 
Antrodiaetus pacificus, and Aliatypus no. 
1). The first method is commonly em- 
ployed by Antrodiaetus unicolor and 
Atypoides riversi, the second method by 
Antrodiaetus pus,nax, Antrodiaetus stygius, 
Aliatypus californicus, and Aliatypus no. 1. 
Sealing is performed by molting spiders, 
some brooding females, and spiders which 
appear to become somewhat inactive dur- 
ing part of the summer. Both sealing 



310 Bulletin Museum of Comparative Zoology, VoJ. 141, No. 6 



methods, particularly the second, probably 
considerably improve the climate control 
and predator and parasite defense func- 
tions hypothesized above for the unsealed 
entrance closure. 

Egg sac structure and position. Egg sac 
structure is similar in all studied species of 
Antrodiaetus and Atypoides, whereas in 
AUafyptis it remains undescribed. Egg sac 
position in the studied species of Antro- 
diaetus and Atypoides (Table 6) often 
varies considerably within species, but in 
all these species, except possibly Atypoides 
gertschi, the egg sac is usually positioned 
between 0.4 and 0.8 of the distance down 
the burrow and usually on the ventral sec- 
tor of the wall. The only species of Alia- 
typus studied (A. no. 1) appears to 
position its egg sac at the very bottom of 
the burrow. Those species found in habi- 
tats with the dryest soil {Antrodiaetus 
montanus, Antrodiaetus pugnax, and Alia- 
typus no. 1 ) appear to place their egg sacs 
deepest in the buiTow, probably because 
of the greater humidity there. 

Brood behavior. In studied species of 
all three genera, the brood emerges from 
the egg sac either in the second instar or 
late in the first instar. There is substantial 
evidence that in Antrodiaetus unicolor the 
second instar brood overwinters in the 
parental burrow and disperses in the 
spring. This might greatly increase the 
winter survival of offspring of species 
living in areas with cold winters. The 
second instar of all studied species is quite 
active, morphologically equipped for most 
adult functions, and, in A. unicolor at least, 
is capable of constmcting burrows. 

Molting behavior. In Antrodiaetus uni- 
color, Antrodiaetus pacificus, and Atypoides 
riversi, molting commonly occurs in sealed 
burrows and is perfonned on the burrow 
wall above the bottom of the burrow. 
After ecdysis the spider rests in the bottom 
end while sclerotization proceeds. Even- 
tually the exuvia is removed from the bur- 
row wall and packed into the bottom end 
of the burrow. Data for other species in 



all three genera appear to fit this scheme. 
Scanty data indicate that Antrodiaetus 
roretzi may behave differently. 

Defensive behavior. When prodded, 
several species of Antrodiaetus and Atypoi- 
des riversi assume an aggressive defensive 
posture similar to that of Antrodiaetus uni- 
color at the bottom end of the burrow. 
Other species of these two genera and 
Aliatypus are unstudied. The bottom end 
of the A. riversi burrow is uniquely nar- 
rowed, resulting in a tighter fit for the 
abdomen when in this defensive posture, 
and is thus probably more effective 
protection. Probably pompilid wasps and 
possibly centipedes are two important 
predators against which such defensive 
behavior is employed. 

Feeding behavior. In the few species of 
Antrodiaetus and Atypoides studied, adults 
forage chiefly at night, and the foraging 
posture of each species appears similar. 
Considerable evidence for Antrodiaetus 
unicolor and Atypoides riversi indicates 
that prey is detected solely by substrate 
vibrations. A. unicolor appears to be an 
unselective predator. Many species com- 
monly pack fragmented prey exoskeletons 
into the bottom end of the burrow. Very 
little data exist on the feeding behavior of 
Aliatyjnis. 

Mating behavior. Evidence from several 
species of Antrodiaetus and Atypoides in- 
dicates that males wander in search of 
mates only at night. Certainly such noc- 
timial behavior would reduce the threat 
from visual predators and desiccation. Pre- 
sumably at least the first stages of court- 
ship or mating also take place then, with 
vision playing no part. Mating behavior 
has been observed only once, in Atypoides 
riversi. However, with this observation, 
with knowledge of antrodiaetid male 
morphology, and with observations on the 
functional morphology of male mating 
structures in other groups of mygalomorph 
spiders, it is possible to form a predictive 
hypothesis about mating behavior in the 
Antrodiaetidae. 



Systematics and Biology of Antrodiaetus 



Coijle 



311 



In Atypoides riversi, the cheliceral 
apoph\'ses of the male are placed between 
the outspread chelicerae of the female to 
hold her in position during mating. The 
similarly \\'ell-de\'eloped male cheliceral 
apophyses of the other Atypoides species 
almost certainly perfonn the same function. 
Perhaps the male first legs aid in holding 
the female, but only A. hadros males 
possess a fairly strongly modified first leg 
(with macrosetae grouped at the proximal 
end of the tibia). 

Males of all species of Antrodiaetus lack 
cheliceral apophyses. (The anterior-dorsal 
prominence on the chelicerae of Antro- 
diaetus males is apparently a vestigial 
cheliceral apophysis.) However, most spe- 
cies possess first leg modifications anal- 
ogous to the tibial spurs and other first leg 
structures, which, in males of many 
mygalomorph groups, function in holding 
the female during mating (for example see 
Buchli, 1962; Petrunkevitch, 1911a; and 
Todd, 1945). In the unicolor species group 
and in Antrodiaetus roretzi, there are 
clusters of strong macrosetae on the pro- 
lateral and, often, ventral surfaces of the 



male first tibia and sometimes a sharp 
bend in the male first metatarsus. Presum- 
ably these structures contact some part of 
the female and hold her in position during 
mating. The male first legs of the lincoln- 
ianus species group are only weakly modi- 
fied, possessing a presumably homologous 
group of enlarged setae on the prolateral 
surface of the first tibia. Probably Antro- 
diaetus males hold the females during 
mating primarily with the modified first 
legs. The weakly modified chelicerae are 
probably less important. 

Males of all Aliatypus species lack 
cheliceral apophyses and possess no special 
holding modifications on their legs. How- 
ever, the pedipalps, unlike those of the 
other two genera, are extremely elongate. 
Perhaps in Aliatypus the male, when 
mating, contacts the female only lightly at 
a distance and reaches the greater distance 
to her genital opening with his long pedi- 
palps. Studies of mating behavior should 
receive high priority among the different 
approaches used to unravel the evolution- 
ary history of the Antrodiaetidae and the 
other atypoid mygalomorph families. 



312 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



Table 1. Timing of brood development in Antrodiaetus unicolor populations. 

(The number of broods collected in a given developmental stage at each locality 

are recorded chronologically by collection date.) 







Large 
















maturing 
















eggs in 


Eggs in 




First 


First 


Second 


Date 


Locality 


abdomen 


egg sac 


Hatching 


instar 


ecdysis 


instar 



April 


1 


Ohio, Adams Co. 


May 


18-19 


W. Va., H 






20 


\y. \'a., Pocahontas Co 


June 


11-27 


Ala., R 




.July 


2 


W. Va., Kal 


'eigh Co. 




6 


Ohio, B 






21-22 


Ky., K 






23-25 


Tenn., O 






27-28 


Ala., R 






29 


Ga., Murray 


' Co. 




31 


Ga., Towns 


Co. 


Aug. 


2-3 


N. C, P 






7-8 


N. C, N 






14-16 


W. Va., C 






21-23 


Va., 7 






25-26 


Va., / 






28-29 


N. C, M 






31 


Tenn., L 




Sept. 


1 


Tenn., O 






5-7 


Ala., Q 






11 


Ark., Stone 


Co. 




14-16 


111., F 






22 


Ohio, B 




Dec. 


2-3 


Ala., Q 





1 

2 
1 

7 



2 
1 



1 
2 
7 
18 
3 



4 
1 

4 
3 



10 

4 
3 
2 



2 
6 



1 
2 
4 
1 



3 
5 
9 
9 
2 
7 
22 
1 
6 
3 
5 



Table 3. Brood size (number of offspring 

PER brood) of ANTRODIAETID SPECIES. 



Table 


2. Brood 










N 
( broods ) 


Range 


Mean 


Std. dev. 


size (number of ( 


offspring 














PER 


brood) of 


Antrodiaetus unicolor 


Antrodiaetus 














population samples. 




unicolor 




87 


38-383 


145.7 


±57.0 












rohustus 
pacificus 




2 
4 


86-94 
107-328 


90.0 
192.2 






N 










Sample 


( broods ) 


Range 


Mean 


Std. dev. 


])u^nax 




8 


40-75 


52.1 


±12.1 


B 


2 


127-155 


141.0 




montanus 




1 


78 






C 


9 


79-185 


120.2 


±36.5 


stygius 




1 


79 






I 


4 


90-209 


127.0 




Atijpoides 












J 


5 


99-318 


189.0 


±87.1 


riversi (coastal) 


7 


43-80 


63.1 


+ 13.5 


K 


6 


84-241 


142.9 


±54.2 


riversi (Sierran) 


1 


81 






M 


1 


102 






gciischi 




1 


78 






N 


2 


111-154 


132.5 




hadros 




1 


80 






O 


31 


38-236 


144.1 


±49.8 














P 


14 


96-256 


169.8 


±52.3 


Aliati/))us 












Q 


7 


66-125 


100.4 


±18.6 


sp. no. 1 




3 


66-104 


85.0 




R 


3 


114-160 


134.6 




sp. no. 2 




1 


139 







Systematics and Biology of Antrodiaetus 



Coylc 



313 



Table 4. Egg size of antrodiaetid species. 
(Ten eggs measured in each brood.) 





N 
(no. of 
broods ) 


Egg diam. 
Range 


in mm 
Mean 


Time 

( months ) 

in 80% 

EtOH 

between 

collecting 

and 
measure- 
ment 


Antrodiaetus 










unicolor 


15 


1.02-1.75 


1.30 


12-28 


pacificus 


3 


1.13-1.46 


1.32 


9 


pugnux 


8 


1.20-1.58 


1.44 


9 


montanus 


1 


1.78-1.95 


1.81 


9 


Atypoides 










riversi 










( coastal ) 


2 


1.54-1.73 


1.63 


5 


riversi 










( Sierran ) 


1 


1.36-1.51 


1.47 


5 



Taule 5. Burrow dimensions and egg sac position of Antrodiaetus unicolor population samples. 
(Only adult female or adult or penultimate male burrow dimensions included. Sample size, range, and 
mean given for each sex of each sample. Burrow length is length in cm of longitudinal axis of burrow 
from ground surface to bottom end. Entrance diameter is maximum inside diameter in mm of bmrow 
entrance opening. Egg sac position is a ratio: length of longitudinal axis of burrow from ground surface 

to upper end of egg sac/burrow length. ) 



Sample 




Burrow length 




Entrance diam. 




Egg sac position 


B 


? 


19 


14-25 


19.6 


18 


11-17 


13.3 


3 


0.13-0.56 


0.28 




S 


2 


16 


16.0 


2 


8-9 


8.5 








C 


2 


21 


15-30 


20.7 


14 


10-18 


13.9 


8 


0.07-0.78 


0.21 




S 


5 


14-24 


18.4 


1 


9 










Chfty Falls, Ind. 


9 


2 


27-38 


32.5 


2 


12-17 


14.5 








D 


9 


9 


11-29 


19.5 


9 


11-20 


14.7 








Crandview St. Pk., W. Va. 


9 


2 


13-19 


16.0 


2 


9-11 


10.0 


1 


0.38 




F 


9 


9 


14-30 


20.3 


6 


10-13 


12.0 










$ 


2 


^ 10 


10.0 


1 


10 










I 


9 


29 


12-30 


20.4 


17 


12-16 


13.8 


7 


0.11-0.67 


0.33 




c? 


9 


6-15 


10.3 


2 


9-10 


9.5 








J 


9 


37 


15-38 


24.6 


30 


12-20 


15.1 


8 


0.05-0.76 


0.34 


K 


9 


14 


14-31 


20.9 


10 


12-17 


14.7 


7 


0.06-0.72 


0.49 


L 


9 


4 
1 


17-36 
4 


23.7 


3 


12-16 


14.3 








M 


9 


20 


10-30 


20.9 


10 


11-15 


12.9 


3 


0.56-0.67 


0.61 


N 


9 


3 


11-18 


15.3 


2 


11-13 


12.0 


2 


0.56-0.73 


0.64 


O 


9 


48 

1 


9-33 
15 


20.6 


45 


11-14 


12.5 


31 


0.48-0.83 


0.66 


P 


9 


35 
6 


10-29 
11-16 


20.9 
14.0 


22 


11-15 


13.5 


15 


0.08-0.75 


0.40 


Q 


9 


35 


10-29 


20.1 


22 


10-14 


12.3 


12 


0.35-0.85 


0.65 


R 


9 


8 


1(^33 


24.0 


8 


11-13 


11.8 


3 


0.48-0.67 


0.59 


Blanchard Springs, Ark. 


9 


2 
1 


11-21 
8 


16.0 















314 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



Table 6. Burrow orviENSiONS and egg sac position of antrodiaetid species. 

(Variables and samples defined as in Talile 5. Turret lieight is length in mm 

of longitudinal axis of turret from ground surface to top of turret. ) 







Burrow length 


Entrance diam. 




Turret height 


Egs. 


' sac position 


Antrodiaetus 






















tinicolor 


$ 307 


9-38(21.2) 


226 


9-20(13.4) 








101 


.05-.85(.51) 




$ 


28 


4-24(12.7) 


6 


8-10( 8.8) 












whiistiis 


9 


5 


18-34(26.8) 


4 


10-1,3(11.3) 








2 


.67-.80(.74) 


pacifictis 


9 


22 
8 


16-44(27.6) 
14-36(24.4) 


IS 
4 


11-13(12.3) 
10-11(10.3) 








3 


.13-.67(.48) 


pti^nax 


5 

S 


13 
2 


9^23(13.8) 
7-9 ( 8.0) 


12 

2 


7-13( 9.0) 
8 ( 8.0) 








8 


.47-.85(.78) 


montamis 


2 


3 


31-46(38.3) 


2 


14 (14.0) 








1 


.71 


hageni 


5 
$ 


2 

2 


22-40(31.0) 

22-32(27.0) 


1 


13 












stygitts 


2 

S 


3 

2 


20 (20.0) 
14-16(15.0) 












2 


.60 (.60) 


roretzi 


2 


? 


10-15 


—from 


Yamamoto (1942) 












S 


2 


6-10( 8.0)- 


—from 


Yamamoto (1942) 










Atijpoidcs 






















riversi ( coastal ) 


2 
S 


11 
5 


15-32(23.8) 
18-21(19.8) 


8 


11-14(12.6) 


10 


13- 


-60(31.8) 


8 


.43-.59(.50) 


riversi (Sierran) 


2 

S 


10 
4 


18-27(21.4) 
8-20(16.5) 


8 


9-14(11.3) 


10 


5- 


-25(14.7) 


1 


.47 


gertschi 


2 


15 


18-36(29.6) 


13 


13-16(14.0) 








1 


.25 


hadros 


2 


5 

2 


8-15(11.2) 

7-8 ( 7.5) 


2 


9-10( 9.5) 












Aliatijpiis 






















californictis 


2 


3 


16-25(19.0)- 


—2 measurements fro 


m S 


mith 


(1908) 






sp. no. 1 


2 

$ 


4 
1 


31-46(38.0) 

23 


4 

1 


11-13(12.0) 

7 








4 


.90-.95(.92) 



Table 7. The instar composition of pre-emeb- 

GENT, emerging, and emerged ( FROM EGG SAC ) 

broods OF Antrodiaetus vnicolor 

POPULATION SAMPLES. 

( Numbers represent number of broods 
in each category in each sample.) 



Table 8. Frequency of brooding females with 

sealed and unsealed burrow entrances in 

Antrodiaetus vnicolor population samples. 

(Asterisk signifies that majority of females had 

collar open or collar was open and spider was 

also in foraging posture. ) 







Brood 
entirely 
within 




Brood 

emerging 

from 


Brood 
entirely 
outside 






Before brood 
emerges from 


After second instar 

spiderlings begin 

emerging from 


Insta 

of 

broo< 


r 


egg sac 




egg sac 




egg sac 




Sample 


egg 


sac 


egg sac 


1 1 


1+2 2 


1 


1+2 


2 


1 1+2 


2 


Unsealed 


Sealed 


Unsealed 


Sealed 


Sample 
















B 




3 




2 


B 














2 


C 




2 




1 


C 


5 


1 


1 


1 






3 


F 






1 


2 


F 














6 


I 








1 


1 


3 






1 


1 


1 


4 


J 




2 


3 


6 


J 


1 


2 


1 


2 


1 




8 


K 


1 


1 






L 














2 


L 






2 




M 








2 






8 


M 








3 


O 
P 


10 
13 




1 

1 








7 


N 
O 


32* 


1 




6 


Q 


6 


1 






3 




24 


P 

Q 


1 

4* 


2 


14* 


3 

4 


Total 


38 


4 


4 


6 


5 


1 


34 


R 


4* 









Systematics and Biology of Antrodiaetus 



Coyle 



315 



Table 9. Records of pompilid wasp predation on Antrodiaetus unicolor. 



LocaliU' 



Date 



Develoimiental stage of wasp 



Egg 


Early 
instar 
larva 


Larva 

in 
cocoon 


Abandoned 
cocoon 


Soil plug 
in burrow 










? 


1 


1 






no 


1 








no 








1 

1 


yes 




1 






yes — 4, no — 4 


1 








yes 


1 








yes 
no 
no 



Summers 


Co. 


, W. Va. 


July 3 


Jefferson 


Co. 


, Ind. 


July 17 


D 






July 20 


R 






July 27 


Towns Co., < 


Ga. 


July 31 


N 






Aug. 7 


C 






Aug. 14- 


J 






Aug. 25 


M 






Aug. 28 


Q 






Sept. 6 


Johnson 


Co., 


Ark. 


Sept. 10 



-16 



Table 10. Prey found in Antrodiaetus 

unicolor burrows. 

( Numbers indicate in how many of 102 burrows 

each prey taxon was found. The great majority of 

these are prey rejectamenta records; very few are 

records of freshly killed prey.) 



Ants (Foniiicidae) . 65 

Beetles (Coleoptera) 64 

Millipedes (Diplopoda) 11 

Flies (Diptera) 4 

Spiders ( Araneae ) 4 

Ilarvestmen (Opiliones) 4 

Sow bugs (Isopoda) 4 

Snails ( Gastropoda ) 3 

Wasps (Vespidae and Scolioidea) 2 

Crickets (Gryllidae) 1 

Caterpillars ( Lepidoptera: Sphingidae) 1 



316 Bulletin Museum of Comparative Zoology, Vol 141, No. 6 



Q 

Z 

O 

I/) 



< 

2 



rrrrn 




\Z 



r 



i 



(M 



^^ LO ^/^ 
O =5 n 



■ u^ o 

: .- 2 ■£ 

; = fe o 



Z3 






O) 

'o 



OJ 



o 
on 



O 



Q 

Z 

O 
< 



< 

2 




CD *(_) 

=3 o 



CO 



en 



O. 



o 




Systematics and Biology of Antrodiaetus • Coijlc 317 



Figure 1. Collection dates of adult male antrodiaetids. Records grouped into Y2 rnonth periods. One unit of vertical scale 
represents one male. Solid block represents adult male collected after leaving its burrow. Two vertical bars represent adult 
mole collected within its burrow before abandoning it. Vertical lines represent adult male collected either before or after 
leaving its burrow; inadequate data accompanies specimen. Empty rectangle represents date on which a male, collected in 
the penultimate instor, molted to maturity in captivity. All records are from examined specimens with the following exceptions 
indicated by superscript numbers: (1) 3 records from W. Ivie (personal communication), (2) June record is holotype collec- 
tion date (Uyemura, 1942), (3) November record and one December record from Yamamoto (1942), (4) Both records from 
Smith (1908). Figure 2. Timing of brood development in ontrodiaetid species. Collection dates and developmental stages of 
broods collected in the field. Records grouped into I/3 month periods. One unit of vertical scale represents one brood. 
Broods in the form of large maturing eggs within abdomen are represented by an empty rectangle, broods consisting of eggs 
in egg sac by vertical lines, hatching broods by crosshatching, first instar broods by stippling, broods during first ecdysis by 2 
vertical bars, and second Instar broods by solid black. Note: (1) All Atypoides riversi broods are from coastal populations 
except for a single Sierran brood collected in the first 1/3 of August. Figure 3. Collection dates of adult males of Antrodiae- 
tus occultus, Antrodiaetus pacificus, and Antrodiaetus pugnox in the Willamette and Umpqua River valleys from Roseburg, 
Ore., north to the Columbia River. Same symbols and scale as in Fig. 1. Records grouped into '/3 month periods. 



318 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 




Figures 4-5. Early instars of Antrodiaetus unicolor. 4. First instar. 4a. Dorsal view. 4b. Chelicera, retrolateral view. 4c. 
Leg I metatarsus and tarsus, retrolateral view. 4d. Spinnerets and anal tubercle, ventral view. 5. Second instar; same views 
and scales as in Fig. 4. Figures 6-15. Antrodiaetus unicolor burrow structure. 6-9. Adult female burrows; P, J, Q, and O 
respectively. 10. Immature burrow, D. 11-14. Burrow entrance. 11-12. Open collar, frontal and sagittal views. 13-14. 
Closed collar, same views. 15. Burrow attacked by pompilid wasp, C. Figures 16-19. Antrod/oetus pacilicus burrow struc- 
ture. 16-17. Adult female burrows; / and U respectively. 18-19. Open and closed collar, sagittal view. (5 cm scale for 
burrows. 10 mm scale for collars.) 



Systematics and Biology of Antrodiaetus • Coyle 319 




Figures 20-23. Adult female burrows of Antrodiaetus spp. 20. A. pugnax, 8. 21. A. montanus; Pocatello, Ida. 22. A. 
hageni; Trail, B. C. 23. A. sfyg/us, paratype. Figures 24-28. Atypoldes riveni. 24-25. Adult female burrows; D and / 
respectively. 26. Open turret, sagittal section. (Modal form for coastal population. Dotted line indicates form of closed tur- 
ret.) 27-28. Mating posture observed at G. (Drawn from field sketch and notes.) 27. Lateral view. 28. View looking down 
burrow. Figures 29-34. Atypoides gerfschi burrow structure. 29-30. Adult female burrows; N and O respectively. 31-32. 
Open collar, frontal and sagittal views. 33-34. Closed collar, same views. Figures 35-37. Atypoides hadros burrow struc- 
ture. 35. Adult female burrow; Feme Clyffe St. Pk., III. 36-37. Collar of same burrow. 36. Closed, frontal view. 37. 
Open, sagittal view. Figures 38-39. Aliatypus no. 1 adult female burrow entrance; frontal and sagittal views with trapdoor 
open. (5 cm scale for burrows. 10 mm scale for entrances.) 



320 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 




Figures 40-55. Antrodiaetid burrow entrances (adult females only). 40-43. Antrodiaetus unicolor. 40-41. J, collar open 
and closed. 42-43. Q, collar open and closed. 44. Anfrod/'aetus pacllicus; U, collar open. 45-46. Antrodiaetus hageni; 
Trail, B. C. 45. Collar fully open. 46. One lateral sector of collar collapsed by author. 47-50. Atypoides riversi. 47. D. 
48. ;. 49. D. 50. /. 51-53. Atypo/'des gerfsch;'. 51. O, collar open. 52-53. N, collar open and closed. 54-55. Alialy- 
pus no. 1; Riverside, Calif. Trapdoor propped open. 



Systematics and Biology of Antrodiaetus 



Coijle 



321 






Figures 56-58. Drawings of Anirodiaetus unicolor burrow construction behavior. Sketched from frames of 16 mm movie film. 
56. Burrowing. Arrows indicate general direction of forces produced by legs and chelicerae against cell wall. 57. Releas- 
ing load of soil onto rim of cell opening. 58. Silk application. 



322 Bulletin Museum of Comporative Zoology, Vol. 141, No. 6 







Systematics and Biology of Antrodiaetus • Coyle 323 



Figures 56-58. Photos (16-mm movie frames) of Antrodiaetus unicolor burrow construction behavior. 56. Burrowing. 57. 
Releasing load of soil onto rim of cell opening. 58. Silk application. Figures 59-62. Antrodiaetus unicolor prey capture 
behavior. 59. Foraging posture (Photo by R. Boll). (Figs. 60-62 are 16-mm movie frames.) 60. Initial contact with prey 
(renebrio larva) at end of lunge from burrow entrance. 61. Fangs imbedded in prey. 62. Spider retreating into burrow 
with prey. 



324 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 




Figures 63-68. Habitat photos. 63. Conifer forest habitat occu pied by Antrodiaetus pacificus in Cascade Mtn. Range near 
L. 64. Riverbank woodland habitat occupied by /Anfrodioe/us pugnax a\ Lewis and Clark St. Pk., Wash. 65. Juniper-soge- 
brush habitat occupied by Antrodiaetus montanus at Pocatello, Ida. 66. Nonforested habitat occupied by Antrodiaetus hageni 
at Trail, B. C. 67. Tan oak-pacific madrone forest habitat occupied by Atypoides riversi a\ D. 68. California red fir habitat 
occupied by Atypoides gertschi at 7,400 ft at N. 



Systematics and Biology of Antrodiaetus 



Coijle 



325 



TAXONOMIC METHODS 
Analysis of Varmtion 

I have approached the species level 
taxonomy of Antrodiaetus with two princi- 
pal and interrelated aims: to discover 
moiphological characters of high value in 
distinguishing between species, and to 
study geographic variation. Difficulty in 
achieving the first goal in many groups of 
mygalomorph spiders (including the Antro- 
diaetidae) is primarily a result of the rel- 
atively simple morphology of the male 
palpus and female genitalia, the difficulty 
of obtaining samples of adequate size, and 
the instar heterogeneity (and therefore 
greater variation) of adult female samples. 
The second difficulty has been overcome 
for many antrodiaetid species and the third 
has been reduced by the use of ratio 
characters. The study of geographic vari- 
ation, also affected by the latter two dif- 
ficulties, is particularly important for 
mygalomorph species, which often exhibit 
much geographic variation within a rel- 
atively small area. Both aims were pur- 
sued by an analysis of variation of 
numerous quantitative and qualitative 
characters. Initially, variation in a large 
number of characters was briefly surveyed, 
and from these the diagnostically most 
promising were selected and their variation 
studied in depth. 

Variation of quantitative characters 
(measurements, meristic characters, and 
ratios fonned from these) was analyzed 
with the aid of a 7094 IBM computer. A 
Fortran II program directed the computer 
to calculate the mean and standard devi- 
ation of each character for each local popu- 
lation sample of each sex and for certain 
groupings of local samples plus indi\'idual 
specimens into larger infraspecific or spe- 
cies samples. The computer then com- 
pared these samples pairwise in any 
desired combination, giving for each char- 
acter for each comparison a value of the 
distinctness of the two samples. This "dis- 
tance" value equals the difference between 



the means of the two samples divided by 
the sum of their standard deviations. I 
could then quickly select those quantitative 
characters of greatest diagnostic value, 
those showing the most marked geographic 
variation, and those infraspecific samples 
that were most divergent. 

This analysis of variation was performed 
on the following number of quantitative 
characters: 14-15 measurements, 6 meristic 
characters, and 35 ratios for males; 17 
measurements, 9 meristic characters, and 
49 ratios for females. These characters 
were defined so as to be well delimited. 
Their definitions and abbreviations are as 
follows (see Figs. 71-75): 
CL Maximum length of carapace (between 

lines tangent to anteriormost and pos- 
teriormost parts of carapace) along line 
parallel to median longitudinal axis, with 
lateral border of carapace horizontal. 
CW Maximum width of carapace along line 

perpendicular to median longitudinal 
axis. 
OQW Maximum widtli of eye group (ocular 
quadrangle) on line perpendicular to 
median longitudinal axis of carapace. 
All eye measurements are made in dorsal 
view with lateral border of carapace 
horizontal. 
ALS Minimum distance between anterior lat- 

eral eyes. 
ALD Maximum diameter of left anterior 

lateral eye. 
AMS Minimum distance between pupils (light 

colored saucer-shaped central area of 
eye) of anterior median eyes. 
AMD Transverse diameter of left anterior 

median eye pupil. 
SL Maximum length of sternum on line 

parallel to median longitudinal axis. 
Anterior border of sternum is its pointed 
anterior extension lateral to labium. 
SW Maximum width of sternum perpen- 

dicular to line defining SL. 
IFL Length of femur I taken as length of 

straight line connecting the proximal and 
distal points of articulation. All leg and 
pedipalp segment length measurements 
were made in side view along retrolateral 
surface of appendanges after removing 
them from spider. 
ITL Length of tibia I taken as length of 

straight line connecting proximal and 
distal points of articulation. 
IML Length of metatarsus I taken as length 



326 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



PTL 




Figure 70. Macrosetae positions on metatarsus I of adult males of Anfrodiaefus. Figures 71-75. Measurements. Figure 76. 
Area around genital opening of Anfrodlaetus un/co/or, showing epiandrous gland spigots. 0.3 mm scale. Figures 77-80. 
Macrosetae types. 71-T&. Ensiform. 79-80. Attenuate. 0.3 mm scale. 



of straight line connecting pro.ximal 
point of articulation with distalniost 
point of segment. 

ITarL Length of tarsus I taken as lengtli of 
straight line connecting most proximal 
exposed point of tarsus with distalniost 
point of dorsal smface. 

I\'FL, I\TL, IVML, IVTarL Leg IV segment 
lengtlis measured in same manner as 
corresponding leg I segments. 

ITD, IMD Diameter of male tibia ( metatarsus ) 
I at ventral swelling in lateral view (A. 
pugnax only ) . 

PFL Length of pedipalpal femur measured 
same as IFL. 

PTL Length of pedipalpal tibia measured 

same as ITL. 

PTT Maximum diameter, taken peipendicular 

to line defining PTL, of pedipalpal tibia 
in lateral view. 

ETL Distance which tip of inner conductor 
sclerite extends beyond tip of outer con- 
ductor sclerite (A. unicolor and A. ro- 
hustus only). 

CT Number of cheliceral macroteeth (per 

chehcera ) . 

CMT Number of cheliceral microteeth (per 
chehcera ) . 

PTSP Number of ensiform macrosetae on pro- 
lateral surface of tarsus of female pedi- 
palp. 

PTSR Number of ensiform macrosetae on 



retrolateral surface of tarsus of female 
pedipalp. 

IMS Number of ensiform macrosetae on 

metatarsus of leg I of female. 

IVMT Number of trichobothria in row on dorsal 
surface of metatarsus IV. As these 
usually vary greatly in size, some care 
must be taken to count small ones also. 

IVMCR Number of setae in large comb on retro- 
lateral surface at distal end of meta- 
tarsus IV. 

IVCTR Number of teeth on retrolateral claw of 
tarsus IV. Claw tooth count includes 
even the reduced proximal teeth. Counts 
must be made carefully under high mag- 
nification in strong light. 

ECS Number of epiandrous gland spigots. 

These are located just anterior to genital 
opening on abdomen of adult males 
(Fig. 76). 

All measurements and counts were per- 
formed by myself with tlie same stereo- 
microscope and eyepiece micrometer scale. 
Five specimens were remeasured five times 
for each character during the course of 
this study; this sampling indicated that the 
measurements are accurate to one microm- 
eter unit for each of the four different 
powers of magnification used. One mi- 



Systematics and Biology of Antrodiaetus 



Coijle 



327 



crometer unit had the following value for 
the following characters: 0.0753 mm for 
CL; 0.0377 mm for CW, SL, SW, and all 
leg and pedipalp segment lengths; 0.0182 
mm for PTT, ITD, and IMD; and 0.0092 
mm for all eye group measurements and 
ETL. 

Many of the quantitative characters 
studied proved to be diagnostically useful. 
Tables 12-14 record for all species the 
variation found in the measurements, me- 
ristic characters, and diagnostically most 
valuable ratios. Male measurements of 
greatest diagnostic value are those of leg I 
segments, pedipalp segments, and anterior 
median eye diameter and separation. 
Ratios formed from these are even more 
useful. Male leg I macrosetae counts are 
also particularly useful. Females are much 
more difficult to diagnose with quantitative 
characters; ratios are usually much more 
useful than nonratio characters. 

Variation in qualitative characters ( those 
not defined strictly quantitativclv) was re- 
corded as carefully as possible and is 
usually described with the aid of figures. 
Color variation of A. unicolor females was 
quantified as follows. The color of the 
lateral slope of the pars cephalica was 
recorded because of the even coloration, 
ease of observation, and rather wide range 
of color variation of this area. All speci- 
mens studied had fully sclerotized fangs 
and had been collected and preserved in 
80 per cent ethanol four to six months 
earlier. The total range of color variation 
was represented by a rectangular grid 
(Fig. 92) of 30 squares with the vertical 
scale representing a basic hue change from 
yellow to orange and the horizontal (gray- 
ness) scale a change in the amount of 
black in the color. The limits of this range 
are pale yellow ( 1 A ) and dark brown 
(5.5C). Ten specimens separated by uni- 
fonu grayness intervals along the A ( lA- 
5A ) and C ( 1C-5C ) rows were selected 
as standards and arranged in a shallow tray 
of alcohol under the even light of two cool 



white F15T8-CW fluorescent bulbs. All 
other specimens were compared individu- 
ally in the tray with these and each was 
assigned a value corresponding to its 
position on the color grid. 

The diagnostically most useful male 
qualitative characters are the shapes and 
macrosetation of tibia I and metatarsus I, 
the pedipalpal tibia shape, and the shapes 
and spatial relationships of the palpus 
sclerites. Females are considerably more 
difficult to separate with qualitative char- 
acters than are males. Cheliceral and 
carapace setation, coloration, cheliceral 
microtooth size and arrangement, and 
seminal receptacle form are the most useful 
female characters in species diagnoses. 

Species sample sizes are given in Tables 
13 and 14, and population sample sizes are 
indicated in the Dice-Leraas diagrams. 
Each locality from which a significant 
sample of A. unicolor, A. pacifictis, or A. 
pugnax was obtained is labeled on the 
maps of Figures 85, 98, and 104 and identi- 
fied in the records sections and in the text 
by an italicized capital letter. 

No external structure or characteristic 
gross change in seminal receptacle fonu 
has been found that indicates when a fe- 
male antrodiaetid has reached sexual 
maturity (or what instar an adult female 
may be). Therefore a female specimen 
was included in a population sample only 
if it had a longer carapace than the 
smallest reproductively active female in 
that sample. Exceptions were made for a 
few rare species for which no reproduc- 
tively active females were available. (See 
species sample sizes given in Table 14.) 
A reproductively active female is defined 
as one with large ripening eggs in her 
abdomen or a brood in her burrow. Other 
females included in a sample therefore 
represent first adult instar females col- 
lected in the summer, fall, or winter before 
or soon after their initial mating, some later 
adult instar females, and occasionally an 
immature female. 



328 BiiUetin Museum of Comparative Zoology, Vol. 141, No. 6 



Material 

This study is based upon the following 
material examined: nearly 370 Antrodioetus 
males (about 250 are A. tinicolor and A. 
pacificus), approximately 500 Anfrodiactus 
females (almost 420 are A. nnicolor and A. 
pacificiis), 28 Alkitijpus males, 55 Aliaty- 
piis females, and roughly 2000 immature 
specimens. Approximately half of this 
material I have collected. The rest was 
loaned to me by the following people and 
institutions: Personal collections — J. A. 
Beatty, J. Carico, P. Craig, H. Exline Friz- 
zell, R. Hoffman, W. Icenogle, R. E. Leech, 
J. MacMahon, B. J. Marples, F. J. Moore, 
P. Nelson, W. Peck, V. D. Roth, W. A. 
Shear, R. Snetzinger, M. Tliompson, B. 
Vogel, A. A. Weaver, T. Yaginuma. Insti- 
tutions — American Museum of Natural 
History (W. J. Gertsch); British Columbia 
Provincial Museum (E. Thorn); British 
Museum (Nat. Hist.) (J. G. Sheals, D. J. 
Clark); California Academy of Science; 
Canadian National Collection (A. L. Tum- 
bull); University of Kansas; Museum of 
Comparative Zoology (H. W. Levi); Uni- 
versity of Nebraska (P. C. Peterson); 
Museum National d'Histoirc Naturelle. 
Paris (M. Vachon, M. Hubert); Peabody 
Museum Natural History, Yale (W. Hart- 
man, C. Remington); and. University of 
Utah (G. Edmunds). 

Collecting Methods 

The uncommonness of antrodiaetids in 
collections is due to their covert behavior. 
Concentration on particular collecting strat- 
egies greatly improves one's chances of 
finding these spiders. Burrow aggregations 
are best discovered by locating promising 
habitats during daylight and searching 
these habitats for open burrow entrances at 
night \\'ith a headlamp. One can then trap 
and collect the spiders at their burrow 
entrances by quickly thrusting a knife 
blade through the soil and across the bur- 
row lumen just below the spider. More 
information is gained by excavating the 



burrow, preferably in daylight after mark- 
ing it at night. As one accumulates ex- 
perience, it becomes possible to efficiently 
locate closed entrances in the daylight by 
carefully examining and delicately picking 
at likely microhabitat surfaces with a knife 
blade. During the mating season, one can 
( with a headlamp ) readily spot adult males 
as thev nocturnallv wander owy the 
ground surface in the vicinity of burrow 
aggregations. Pitfall traps are excellent 
collectors of wandering males. Careful 
searching for sealed or nocturnally closed 
burrow entrances early in the mating 
season will often reveal penultimate males 
and recently molted adult males. Penulti- 
mate males, easily recognized by their 
swollen pedipalpal tarsi (Coyle, 196(S, fig. 
45), will frequently molt to adulthood if 
maintained in a cool, humid, and dark en- 
vironment. 

Morphological Terminology 

Eyes. The abbreviations AME, ALE, 
PME, and PLE designate the following 
eyes respectively: anterior median eyes, 
anterior lateral eyes, posterior median eyes, 
and posterior lateral eyes. 

Chcliccrae. A variable prominence, prob- 
ably homologous to the chcliccral apophysis 
of Atypoides, is found on the anterior-dorsal 
surface of the chelicerae of Antrodioetus 
males (Figs. 145-157). This is referred to 
as the anterior-dorsal eheliccral promi- 
nence. The clieliceral macroteeth are the 
large teeth forming a single prolateral row 
(Fig. 158) (and a second more retrolateral 
row in Aliatypus). The chelicerol micro- 
teeth are the much smaller teeth scattered 
retrolaterally along part of the prolateral 
macrotooth row. In the descriptions of 
microteeth position, the tenn "last" means 
most proximal. The presence or absence of 
a setaless area on the upper ectal (=retro- 
lateral) surface of the chelicera is a useful 
character in the nnicolor group of Antro- 
diaetus (Figs. 170-172). The always seta- 
less ectal and dorsal surface proximal of 



Systematics and Biology of Antrodiaetvs 



Coyle 



329 



the dorsal gibbosity is not part of this 
character. 

Polpus. In Antrodiaetiis, as in Atypoides 
and Aliatypus, the conductor of the palpus 
consists of two sclerites (Fig. 207), an 
inner conductor sclerite (i.c.s.), which 
tapers and at least partly envelops the 
embolus distally, and an outer conductor 
sclerite (o.c.s.), which lies outside and 
cradles these structures. 

Male leg I. An ensiforin macroseta is 
one which tapers abruptly at its terminal 
end and is therefore rigid for its entire 
length (Figs. 77, 78). An attenuate macro- 
seta tapers gradually and is therefore very 
slender distally and often easily bent (Figs. 
79, SO). At least 90 per cent of all male 
til^a I macrosetae can be definitely as- 
signed to one or the other of these two 
categories. Attenuate macrosetae with 
broken tips are readily recognized if one 
is cautious. The macrosetae pattern of 
male metatarsus I is diagnostically quite 
useful. Six such differently positioned 
macrosetae occur in Antrodiaetiis and are 
identified by the letters A through F (Fig. 
70). When present, each is in nearly the 
same position in every species and there- 
fore probably homologous throughout the 
genus, i.e., macroseta A of A. nnicolor is 
homologous to that of A. pugnax, etc. 
However, macroseta B shows some position 
variation, but it seems to vary as much 
within A. pugnax as between other species. 
In A. unicolor, A. rnontanus, A. stygitis, and 
A. apachecus on most male metatarsi I 
which are recorded as lacking macrosetae 
in Table 12, macroseta A is represented by 
a rather prominent enlarged seta. 

Abdomen. The abbreviations AL, PM, 
and PL designate the following spinnerets 
respectively: anterior lateral, posterior 
median, and posterior lateral. The anterior 
half of the abdominal dorsum of Antro- 
diaetus is provided with constantly posi- 
tioned, segmentally arranged patches of 
more heavily sclerotized cuticle (Fig. 109), 
presumably vestigial abdominal tergites. 
These may be continuous, i.e., fused to one 



another at adjacent borders. The second 
patch (from the anterior end), which is 
always present in both sexes, is smooth, 
platelike, and most heavily sclerotized; it 
is called a tergite in this paper. 

Female genitalia. The general mor- 
phology of the female genitalia in Antro- 
diaettis ( Fig. 275 ) is like that of Atypoides 
(see Coyle, 1968). 

Methods of Presentation 

Descriptions. Only the most important 
literature references are cited. Complete 
citations are given in Coyle, 1969. Each 
description is a composite of all adult 
specimens at hand. Any characters in 
which the holotype or lectotype is mark- 
edly variant are noted. Only characters of 
diagnostic value are included. The quanti- 
tative characters recorded in Tables 12-15 
are an integral part of each description. 
Colors are described from fully sclerotized 
specimens immersed in ethanol under 
strong fluorescent light. Only specimens 
which have been dead in ethanol from six 
months to five years were used for color 
descriptions, with a few exceptions as 
noted. 

Diagnosis. The diagnosis lists characters 
most useful in identifying a given species. 
These characters are Hsted in the approxi- 
mate order of their diagnostic value. Since 
I often have not repeated characters in 
the diagnoses of all of several (or a pair 
of) closely related species, one should 
examine the diagnoses of all these possi- 
bilities when attempting a difficult identi- 
fication. The absence of female characters 
with high diagnostic value in many species 
calls for extreme care when identifying 
females. It must also be kept in mind, 
when using a diagnosis based upon a small 
sample, that the known variation range is 
probably considerably less than that of the 
whole species population. 

Illustrations. Illustrations were carefully 
constructed on translucent paper over a 
squared grid template with the aid of a 
squared grid reticle in the eyepiece of the 



330 Bulletin Museum of Comparative Zoology, Vol. 141, No. 6 



stereomicroscope. Figures of female gen- 
italia are always drawn from reprodnc- 
tively active females unless otherwise 
noted. The thin-walled ducts leading from 
the seminal receptacle bases to the bursa 
copulatrix roof are usually incompletely 
drawn; their openings into the bursa 
copulatrix are difficult to see under nomial 
illumination. 

Records. Unless otherwise noted, only 
specimens that I have examined are listed 
in the records section. States and counties 
(and Japanese prefectures) are given in 
alphabetical order. Within each county 
citation, all records from a given locality 
are separated from those of other localities 
by a dash. Collection dates are listed only 
for males. "( $ )" means that the specimen 
was collected as a penultimate instar on 
the date indicated and matured later in 
captivity. When no "S" or "9" follows a 
record, this means that only immatures 
were collected, 

TAXONOMY 

ANTRODIAETIDAE Gertsch, 1940 

Brachybothriinae Simon, 1892, Histoire Naturelle 

des Araignees, 1(1): 193. Type genus Brachij- 

bothriiim Simon, 1884 {=Antwdiaetiis Aus- 

serer, 1871). 
Brachybothriidae Pocock, 1903, Proc. Zool. Soc. 

London, (1): 346. 
Acattymidae Kisbida, 1930, Lansania, 2(13): 34. 

Type genus Acatttjma L. Kocb, 1878 ( =A»if/o- 

diaetus Ausserer, 1871). 
Antrodiaetinae Gertsch, 1940, in Comstock (rev. 

ed.). The Spider Book, p. 236. Type genus 

Antrodiaetiis Ausserer, 1871. 
Antrodiaetidae Kaston, 1948, Connecticut Geol. 

Nat. Hist. Surv. Bull., (70): 48. 

Note on family names. Although the 
name Brachybothriidae is older than either 
Acattymidae or Antrodiaetidae, Antrodiae- 
tidae has clearly won general acceptance. 
Antrodiaetidae (and -inae) has been used 
more frequently (in approximately 17 
publications) than cither Brachybothriidae 
(and -inae) (in approximately 14 publi- 
cations) or Acattymidae (in approximately 
7 publications). Antrodiaetidae is the only 
name used since 1957 (in approximately 



14 publications). It is the only name used 
in works presently reaching a relatively 
large audience of both professionals and 
amateurs (Comstock, 1940; Kaston, 1948; 
Kaston and Kaston, 1953; Yaginuma, 1960; 
Kaestner, 1968; Levi and Levi, 1968), with 
the exception of Gertsch's ( 1949 ) use of 
Acattymidae. In accordance with Article 
40 of the recent ( 1961 ) International Code 
of Zoological Nomenclature, both the pres- 
ent general acceptance of the name Antro- 
diaetidae and the replacement before 1961 
of the two older names by Antrodiaetidae 
because of the synonymy of their type 
genera with the senior synonym Antro- 
diaetiis (Kaston, 1948; Yaginuma, 1962) 
clearly justify rejection of the older names 
and adoption of Antrodiaetidae. 

Characteristics. Carapace: Pars cephalica 
elevated above pars thoracica. Eyes 
grouped on a median prominence. ALE's 
largest; fonu a slightly procurved trans- 
verse row with AME's. PLE's form lateral 
limits of eye group. PME's widely sepa- 
rated; close to respective PLE's. Sternum: 
Four pairs of sigilla; anterior pair large, 
just behind labium and sometimes in- 
distinct; posterior pair larger than second 
or third pairs. Labium well defined but 
fused to sternum; inclined from plane of 
sternum. Chelicerae: Female chelicerae 
very robust; strongly gibbous dorsally. 
Rastellum well developed on females. One 
row of macroteeth on prolateral side of 
closed fang; sometimes a macrotooth row 
also on retrolateral side of closed fang. 
Microteeth on retrolateral side of prolateral 
macrotooth row. Pedipalps: Coxal endite 
very small. Female tarsus with 2 rows of 
macrosetae (on prolateral and retrolateral 
aspects of ventral surface). Male tibia 
swollen. Palpus with well defined o.c.s., 
i.c.s., and embolus. I.c.s. at least partly 
envelops embolus distally. Le^s: Female 
legs relatively stout and with many macro- 
setae. Abdomen: Males with 1 to 4 seg- 
mentally arranged (sometimes fused) 
sclerotized patches dorsally; second patch 
most heavily sclerotized, tergitclike, and 



Systematics and Biology of Antrodiaetus • Coijle 



331 



always present. Females possess only this 
second tergite; rarely other patches also. 
Epiandrous gland spigots clustered just 
anterior to male genital opening. AL spin- 
nerets 2-segmented and functional; re- 
duced and unsegmented; or absent. PM's 
unsegmented and functional. PL's 3- 
segmented and functional; distal segment 
length not over li that of middle segment. 
Female (genitalia: Genital opening broad 
slit shared by uterus exit; not markedly 
sclerotized externally. Four seminal recep- 
tacles with stalk and apical enlarged 
portion open into chamber (bursa copula- 
trix) just inside genital opening. 

Diaiinosis. Males of this family can be 
separated from those of the Mecicobothrii- 
dae and Atypidae by the possession of a 
strongly sclerotized i.c.s. which is distinct 
from the o.c.s.; or, if the i.c.s. is not 
strongly sclerotized, then by the possession 
of either only two pairs of spinnerets or 
a pedipalpal patella almost as long as the 
tibia. Antrodiaetid females can be distin- 
guished from both mecicobothriid and 
atypid females by the presence of a ra- 
stellum, from the former group by the 
strongly elevated pars cephalica and a 
proportionately much shorter distal PL 
spinneret segment, and from atypids by 
the two rows of pedipalpal tarsus macro- 
setae. 

Provisional inchision of Aliatypus. As is 
emphasized in the discussion of phylogeny, 
Antrodiaetus and Atypoides form a mono- 
phyletic group distinct from Aliatypus. A 
number of character states which these 
two distinct groups share could be the 
result of convergence or at least parallel 
evolution rather than indicators of recent 
common ancestry. In heavily weighted 
genitalia characters, Aliatypus strongly re- 
sembles the family Mecicobothriidae; it is 
possible that Aliatypus is more closely 
related to this family than to Antrodiaetus 
and Atypoides. Perhaps Aliatypus will 
prove sufficiently distinct phylogenetically 
and ecologically to merit consideration as 
a separate family. ( The resulting four small 



relict atypoid taxa [Liphistiidae not in- 
cluded] might for convenience's sake be 
reduced to subfamilies under a single 
family name.) Only after close study of 
Aliatypus and the mecicobothriids can the 
approximate phylogenetic position and 
proper classification of Aliatypus be re- 
solved. I shall provisionally retain Alia- 
typus within the family Antrodiaetidae. 

Key to Genera of Antrodiaetidae 

la. Thoracic groove longitudinal. AL spin- 
nerets absent (Fig. 318) or unsegmented 
with at most one spigot apically. No che- 
liceral macroteetli on retrolateral side of 
closed fang (Fig. 158). Female IVMT 
= 5-21. Male pedipalpal patell